Merge branch 'master' into feature_weaponscale

.github/workflows/continuous_integration.yml

@@ -11,13 +11,19 @@ jobs:
       matrix:
         config:
         - {
-            name: "Visual Studio 64-bit",
+            name: "Visual Studio 2022",
+            os: windows-2022,
+            extra_options: "-A x64",
+            build_type: "Release"
+          }
+        - {
+            name: "Visual Studio 2019",
             os: windows-2019,
             extra_options: "-A x64",
             build_type: "Release"
           }
         - {
-            name: "Visual Studio 64-bit",
+            name: "Visual Studio 2019",
             os: windows-2019,
             extra_options: "-A x64",
             build_type: "Debug"

libraries/lzma/C/7z.h

@@ -1,5 +1,5 @@
 /* 7z.h -- 7z interface
-2017-04-03 : Igor Pavlov : Public domain */
+2018-07-02 : Igor Pavlov : Public domain */
 
 #ifndef __7Z_H
 #define __7Z_H
@@ -91,6 +91,8 @@ typedef struct
   UInt64 *CoderUnpackSizes;       // for all coders in all folders
 
   Byte *CodersData;
+
+  UInt64 RangeLimit;
 } CSzAr;
 
 UInt64 SzAr_GetFolderUnpackSize(const CSzAr *p, UInt32 folderIndex);

libraries/lzma/C/7zArcIn.c

@@ -1,5 +1,5 @@
 /* 7zArcIn.c -- 7z Input functions
-2018-12-31 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -75,7 +75,7 @@ static SRes SzBitUi32s_Alloc(CSzBitUi32s *p, size_t num, ISzAllocPtr alloc)
   return SZ_OK;
 }
 
-void SzBitUi32s_Free(CSzBitUi32s *p, ISzAllocPtr alloc)
+static void SzBitUi32s_Free(CSzBitUi32s *p, ISzAllocPtr alloc)
 {
   ISzAlloc_Free(alloc, p->Defs); p->Defs = NULL;
   ISzAlloc_Free(alloc, p->Vals); p->Vals = NULL;
@@ -83,7 +83,7 @@ void SzBitUi32s_Free(CSzBitUi32s *p, ISzAllocPtr alloc)
 
 #define SzBitUi64s_Init(p) { (p)->Defs = NULL; (p)->Vals = NULL; }
 
-void SzBitUi64s_Free(CSzBitUi64s *p, ISzAllocPtr alloc)
+static void SzBitUi64s_Free(CSzBitUi64s *p, ISzAllocPtr alloc)
 {
   ISzAlloc_Free(alloc, p->Defs); p->Defs = NULL;
   ISzAlloc_Free(alloc, p->Vals); p->Vals = NULL;
@@ -105,6 +105,8 @@ static void SzAr_Init(CSzAr *p)
   p->CoderUnpackSizes = NULL;
 
   p->CodersData = NULL;
+
+  p->RangeLimit = 0;
 }
 
 static void SzAr_Free(CSzAr *p, ISzAllocPtr alloc)
@@ -502,7 +504,7 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd)
         return SZ_ERROR_ARCHIVE;
       if (propsSize >= 0x80)
         return SZ_ERROR_UNSUPPORTED;
-      coder->PropsOffset = sd->Data - dataStart;
+      coder->PropsOffset = (size_t)(sd->Data - dataStart);
       coder->PropsSize = (Byte)propsSize;
       sd->Data += (size_t)propsSize;
       sd->Size -= (size_t)propsSize;
@@ -677,7 +679,7 @@ static SRes ReadUnpackInfo(CSzAr *p,
   {
     UInt32 numCoders, ci, numInStreams = 0;
     
-    p->FoCodersOffsets[fo] = sd.Data - startBufPtr;
+    p->FoCodersOffsets[fo] = (size_t)(sd.Data - startBufPtr);
     
     RINOK(SzReadNumber32(&sd, &numCoders));
     if (numCoders == 0 || numCoders > k_Scan_NumCoders_MAX)
@@ -797,7 +799,7 @@ static SRes ReadUnpackInfo(CSzAr *p,
   p->FoToCoderUnpackSizes[fo] = numCodersOutStreams;
   
   {
-    size_t dataSize = sd.Data - startBufPtr;
+    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);
@@ -885,7 +887,7 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi)
         if (numStreams != 1 || !SzBitWithVals_Check(&p->FolderCRCs, i))
           numSubDigests += numStreams;
       }
-      ssi->sdNumSubStreams.Size = sd->Data - ssi->sdNumSubStreams.Data;
+      ssi->sdNumSubStreams.Size = (size_t)(sd->Data - ssi->sdNumSubStreams.Data);
       continue;
     }
     if (type == k7zIdCRC || type == k7zIdSize || type == k7zIdEnd)
@@ -907,7 +909,7 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi)
   {
     ssi->sdSizes.Data = sd->Data;
     RINOK(SkipNumbers(sd, numUnpackSizesInData));
-    ssi->sdSizes.Size = sd->Data - ssi->sdSizes.Data;
+    ssi->sdSizes.Size = (size_t)(sd->Data - ssi->sdSizes.Data);
     RINOK(ReadID(sd, &type));
   }
 
@@ -919,7 +921,7 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi)
     {
       ssi->sdCRCs.Data = sd->Data;
       RINOK(SkipBitUi32s(sd, numSubDigests));
-      ssi->sdCRCs.Size = sd->Data - ssi->sdCRCs.Data;
+      ssi->sdCRCs.Size = (size_t)(sd->Data - ssi->sdCRCs.Data);
     }
     else
     {
@@ -947,7 +949,11 @@ static SRes SzReadStreamsInfo(CSzAr *p,
   if (type == k7zIdPackInfo)
   {
     RINOK(ReadNumber(sd, dataOffset));
+    if (*dataOffset > p->RangeLimit)
+      return SZ_ERROR_ARCHIVE;
     RINOK(ReadPackInfo(p, sd, alloc));
+    if (p->PackPositions[p->NumPackStreams] > p->RangeLimit - *dataOffset)
+      return SZ_ERROR_ARCHIVE;
     RINOK(ReadID(sd, &type));
   }
   if (type == k7zIdUnpackInfo)
@@ -1028,12 +1034,12 @@ static SRes SzReadFileNames(const Byte *data, size_t size, UInt32 numFiles, size
       return SZ_ERROR_ARCHIVE;
     for (p = data + pos;
       #ifdef _WIN32
-      *(const UInt16 *)p != 0
+      *(const UInt16 *)(const void *)p != 0
       #else
       p[0] != 0 || p[1] != 0
       #endif
       ; p += 2);
-    pos = p - data + 2;
+    pos = (size_t)(p - data) + 2;
     *offsets++ = (pos >> 1);
   }
   while (--numFiles);
@@ -1133,6 +1139,8 @@ static SRes SzReadHeader2(
     SRes res;
     
     SzAr_Init(&tempAr);
+    tempAr.RangeLimit = p->db.RangeLimit;
+
     res = SzReadAndDecodePackedStreams(inStream, sd, tempBufs, NUM_ADDITIONAL_STREAMS_MAX,
         p->startPosAfterHeader, &tempAr, allocTemp);
     *numTempBufs = tempAr.NumFolders;
@@ -1526,11 +1534,13 @@ static SRes SzArEx_Open2(
   nextHeaderSize = GetUi64(header + 20);
   nextHeaderCRC = GetUi32(header + 28);
 
-  p->startPosAfterHeader = startArcPos + k7zStartHeaderSize;
+  p->startPosAfterHeader = (UInt64)startArcPos + k7zStartHeaderSize;
   
   if (CrcCalc(header + 12, 20) != GetUi32(header + 8))
     return SZ_ERROR_CRC;
 
+  p->db.RangeLimit = nextHeaderOffset;
+
   nextHeaderSizeT = (size_t)nextHeaderSize;
   if (nextHeaderSizeT != nextHeaderSize)
     return SZ_ERROR_MEM;
@@ -1543,13 +1553,13 @@ static SRes SzArEx_Open2(
   {
     Int64 pos = 0;
     RINOK(ILookInStream_Seek(inStream, &pos, SZ_SEEK_END));
-    if ((UInt64)pos < startArcPos + nextHeaderOffset ||
+    if ((UInt64)pos < (UInt64)startArcPos + nextHeaderOffset ||
-        (UInt64)pos < startArcPos + k7zStartHeaderSize + nextHeaderOffset ||
+        (UInt64)pos < (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset ||
-        (UInt64)pos < startArcPos + k7zStartHeaderSize + nextHeaderOffset + nextHeaderSize)
+        (UInt64)pos < (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset + nextHeaderSize)
       return SZ_ERROR_INPUT_EOF;
   }
 
-  RINOK(LookInStream_SeekTo(inStream, startArcPos + k7zStartHeaderSize + nextHeaderOffset));
+  RINOK(LookInStream_SeekTo(inStream, (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset));
 
   if (!Buf_Create(&buf, nextHeaderSizeT, allocTemp))
     return SZ_ERROR_MEM;
@@ -1575,6 +1585,8 @@ static SRes SzArEx_Open2(
         Buf_Init(&tempBuf);
         
         SzAr_Init(&tempAr);
+        tempAr.RangeLimit = p->db.RangeLimit;
+
         res = SzReadAndDecodePackedStreams(inStream, &sd, &tempBuf, 1, p->startPosAfterHeader, &tempAr, allocTemp);
         SzAr_Free(&tempAr, allocTemp);
        

libraries/lzma/C/7zCrc.c

@@ -1,5 +1,5 @@
 /* 7zCrc.c -- CRC32 init
-2017-06-06 : Igor Pavlov : Public domain */
+2021-04-01 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -26,8 +26,20 @@
 
 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;
+extern
+CRC_FUNC g_CrcUpdateT0_32;
+CRC_FUNC g_CrcUpdateT0_32;
+extern
+CRC_FUNC g_CrcUpdateT0_64;
+CRC_FUNC g_CrcUpdateT0_64;
+extern
+CRC_FUNC g_CrcUpdate;
 CRC_FUNC g_CrcUpdate;
 
 UInt32 g_CrcTable[256 * CRC_NUM_TABLES];
@@ -44,6 +56,7 @@ UInt32 MY_FAST_CALL CrcCalc(const void *data, size_t size)
 
 #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)
 {
   const Byte *p = (const Byte *)data;
@@ -53,6 +66,166 @@ UInt32 MY_FAST_CALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const U
   return v;
 }
 
+
+/* ---------- hardware CRC ---------- */
+
+#ifdef MY_CPU_LE
+
+#if defined(MY_CPU_ARM_OR_ARM64)
+
+// #pragma message("ARM*")
+
+  #if defined(_MSC_VER)
+    #if defined(MY_CPU_ARM64)
+    #if (_MSC_VER >= 1910)
+        #define USE_ARM64_CRC
+    #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
+            #define ATTRIB_CRC __attribute__((__target__("arch=armv8-a+crc")))
+          #endif
+      #endif
+      #if defined(__ARM_FEATURE_CRC32)
+        #define USE_ARM64_CRC
+        #include <arm_acle.h>
+      #endif
+  #endif
+
+#else
+
+// no hardware CRC
+
+// #define USE_CRC_EMU
+
+#ifdef USE_CRC_EMU
+
+#pragma message("ARM64 CRC emulation")
+
+MY_FORCE_INLINE
+UInt32 __crc32b(UInt32 v, UInt32 data)
+{
+  const UInt32 *table = g_CrcTable;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data);
+  return v;
+}
+
+MY_FORCE_INLINE
+UInt32 __crc32w(UInt32 v, UInt32 data)
+{
+  const UInt32 *table = g_CrcTable;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  return v;
+}
+
+MY_FORCE_INLINE
+UInt32 __crc32d(UInt32 v, UInt64 data)
+{
+  const UInt32 *table = g_CrcTable;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  v = CRC_UPDATE_BYTE_2(v, (Byte)data); data >>= 8;
+  return v;
+}
+
+#endif // USE_CRC_EMU
+
+#endif // defined(MY_CPU_ARM64) && defined(MY_CPU_LE)
+
+
+
+#if defined(USE_ARM64_CRC) || defined(USE_CRC_EMU)
+
+#define T0_32_UNROLL_BYTES (4 * 4)
+#define T0_64_UNROLL_BYTES (4 * 8)
+
+#ifndef ATTRIB_CRC
+#define ATTRIB_CRC
+#endif
+// #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);
+ATTRIB_CRC
+UInt32 MY_FAST_CALL CrcUpdateT0_32(UInt32 v, const void *data, size_t size, const UInt32 *table)
+{
+  const Byte *p = (const Byte *)data;
+  UNUSED_VAR(table);
+
+  for (; size != 0 && ((unsigned)(ptrdiff_t)p & (T0_32_UNROLL_BYTES - 1)) != 0; size--)
+    v = __crc32b(v, *p++);
+
+  if (size >= T0_32_UNROLL_BYTES)
+  {
+    const Byte *lim = p + size;
+    size &= (T0_32_UNROLL_BYTES - 1);
+    lim -= size;
+    do
+    {
+      v = __crc32w(v, *(const UInt32 *)(const void *)(p));
+      v = __crc32w(v, *(const UInt32 *)(const void *)(p + 4)); p += 2 * 4;
+      v = __crc32w(v, *(const UInt32 *)(const void *)(p));
+      v = __crc32w(v, *(const UInt32 *)(const void *)(p + 4)); p += 2 * 4;
+    }
+    while (p != lim);
+  }
+  
+  for (; size != 0; size--)
+    v = __crc32b(v, *p++);
+
+  return v;
+}
+
+ATTRIB_CRC
+UInt32 MY_FAST_CALL 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)
+{
+  const Byte *p = (const Byte *)data;
+  UNUSED_VAR(table);
+
+  for (; size != 0 && ((unsigned)(ptrdiff_t)p & (T0_64_UNROLL_BYTES - 1)) != 0; size--)
+    v = __crc32b(v, *p++);
+
+  if (size >= T0_64_UNROLL_BYTES)
+  {
+    const Byte *lim = p + size;
+    size &= (T0_64_UNROLL_BYTES - 1);
+    lim -= size;
+    do
+    {
+      v = __crc32d(v, *(const UInt64 *)(const void *)(p));
+      v = __crc32d(v, *(const UInt64 *)(const void *)(p + 8)); p += 2 * 8;
+      v = __crc32d(v, *(const UInt64 *)(const void *)(p));
+      v = __crc32d(v, *(const UInt64 *)(const void *)(p + 8)); p += 2 * 8;
+    }
+    while (p != lim);
+  }
+  
+  for (; size != 0; size--)
+    v = __crc32b(v, *p++);
+
+  return v;
+}
+
+#endif // defined(USE_ARM64_CRC) || defined(USE_CRC_EMU)
+
+#endif // MY_CPU_LE
+
+
+
+
 void MY_FAST_CALL CrcGenerateTable()
 {
   UInt32 i;
@@ -123,6 +296,27 @@ void MY_FAST_CALL CrcGenerateTable()
     }
   }
   #endif
+  #endif
 
+  #ifdef MY_CPU_LE
+    #ifdef USE_ARM64_CRC
+      if (CPU_IsSupported_CRC32())
+      {
+        g_CrcUpdateT0_32 = CrcUpdateT0_32;
+        g_CrcUpdateT0_64 = CrcUpdateT0_64;
+        g_CrcUpdate =
+          #if defined(MY_CPU_ARM)
+            CrcUpdateT0_32;
+          #else
+            CrcUpdateT0_64;
+          #endif
+      }
+    #endif
+    
+    #ifdef USE_CRC_EMU
+      g_CrcUpdateT0_32 = CrcUpdateT0_32;
+      g_CrcUpdateT0_64 = CrcUpdateT0_64;
+      g_CrcUpdate = CrcUpdateT0_64;
+    #endif
   #endif
 }

libraries/lzma/C/7zCrcOpt.c

@@ -1,5 +1,5 @@
 /* 7zCrcOpt.c -- CRC32 calculation
-2017-04-03 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -9,6 +9,7 @@
 
 #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)
 {
   const Byte *p = (const Byte *)data;
@@ -16,7 +17,7 @@ UInt32 MY_FAST_CALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const U
     v = CRC_UPDATE_BYTE_2(v, *p);
   for (; size >= 4; size -= 4, p += 4)
   {
-    v ^= *(const UInt32 *)p;
+    v ^= *(const UInt32 *)(const void *)p;
     v =
           (table + 0x300)[((v      ) & 0xFF)]
         ^ (table + 0x200)[((v >>  8) & 0xFF)]
@@ -28,6 +29,7 @@ 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)
 {
   const Byte *p = (const Byte *)data;
@@ -36,13 +38,13 @@ UInt32 MY_FAST_CALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const U
   for (; size >= 8; size -= 8, p += 8)
   {
     UInt32 d;
-    v ^= *(const UInt32 *)p;
+    v ^= *(const UInt32 *)(const void *)p;
     v =
           (table + 0x700)[((v      ) & 0xFF)]
         ^ (table + 0x600)[((v >>  8) & 0xFF)]
         ^ (table + 0x500)[((v >> 16) & 0xFF)]
         ^ (table + 0x400)[((v >> 24))];
-    d = *((const UInt32 *)p + 1);
+    d = *((const UInt32 *)(const void *)p + 1);
     v ^=
           (table + 0x300)[((d      ) & 0xFF)]
         ^ (table + 0x200)[((d >>  8) & 0xFF)]
@@ -72,7 +74,7 @@ UInt32 MY_FAST_CALL CrcUpdateT1_BeT4(UInt32 v, const void *data, size_t size, co
     v = CRC_UPDATE_BYTE_2_BE(v, *p);
   for (; size >= 4; size -= 4, p += 4)
   {
-    v ^= *(const UInt32 *)p;
+    v ^= *(const UInt32 *)(const void *)p;
     v =
           (table + 0x000)[((v      ) & 0xFF)]
         ^ (table + 0x100)[((v >>  8) & 0xFF)]
@@ -94,13 +96,13 @@ UInt32 MY_FAST_CALL CrcUpdateT1_BeT8(UInt32 v, const void *data, size_t size, co
   for (; size >= 8; size -= 8, p += 8)
   {
     UInt32 d;
-    v ^= *(const UInt32 *)p;
+    v ^= *(const UInt32 *)(const void *)p;
     v =
           (table + 0x400)[((v      ) & 0xFF)]
         ^ (table + 0x500)[((v >>  8) & 0xFF)]
         ^ (table + 0x600)[((v >> 16) & 0xFF)]
         ^ (table + 0x700)[((v >> 24))];
-    d = *((const UInt32 *)p + 1);
+    d = *((const UInt32 *)(const void *)p + 1);
     v ^=
           (table + 0x000)[((d      ) & 0xFF)]
         ^ (table + 0x100)[((d >>  8) & 0xFF)]

libraries/lzma/C/7zDec.c

@@ -1,5 +1,5 @@
 /* 7zDec.c -- Decoding from 7z folder
-2019-02-02 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -21,17 +21,20 @@
 #endif
 
 #define k_Copy 0
-#define k_Delta 3
+#ifndef _7Z_NO_METHOD_LZMA2
 #define k_LZMA2 0x21
+#endif
 #define k_LZMA  0x30101
-#define k_BCJ   0x3030103
 #define k_BCJ2  0x303011B
+#ifndef _7Z_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
 
@@ -56,7 +59,7 @@ static Byte ReadByte(const IByteIn *pp)
     return *p->cur++;
   if (p->res == SZ_OK)
   {
-    size_t size = p->cur - p->begin;
+    size_t size = (size_t)(p->cur - p->begin);
     p->processed += size;
     p->res = ILookInStream_Skip(p->inStream, size);
     size = (1 << 25);
@@ -101,28 +104,32 @@ static SRes SzDecodePpmd(const Byte *props, unsigned propsSize, UInt64 inSize, c
     Ppmd7_Init(&ppmd, order);
   }
   {
-    CPpmd7z_RangeDec rc;
+    ppmd.rc.dec.Stream = &s.vt;
-    Ppmd7z_RangeDec_CreateVTable(&rc);
+    if (!Ppmd7z_RangeDec_Init(&ppmd.rc.dec))
-    rc.Stream = &s.vt;
-    if (!Ppmd7z_RangeDec_Init(&rc))
       res = SZ_ERROR_DATA;
-    else if (s.extra)
+    else if (!s.extra)
-      res = (s.res != SZ_OK ? s.res : SZ_ERROR_DATA);
-    else
     {
-      SizeT i;
+      Byte *buf = outBuffer;
-      for (i = 0; i < outSize; i++)
+      const Byte *lim = buf + outSize;
+      for (; buf != lim; buf++)
       {
-        int sym = Ppmd7_DecodeSymbol(&ppmd, &rc.vt);
+        int sym = Ppmd7z_DecodeSymbol(&ppmd);
         if (s.extra || sym < 0)
           break;
-        outBuffer[i] = (Byte)sym;
+        *buf = (Byte)sym;
       }
-      if (i != outSize)
+      if (buf != lim)
-        res = (s.res != SZ_OK ? s.res : SZ_ERROR_DATA);
-      else if (s.processed + (s.cur - s.begin) != inSize || !Ppmd7z_RangeDec_IsFinishedOK(&rc))
         res = SZ_ERROR_DATA;
+      else if (!Ppmd7z_RangeDec_IsFinishedOK(&ppmd.rc.dec))
+      {
+        /* if (Ppmd7z_DecodeSymbol(&ppmd) != PPMD7_SYM_END || !Ppmd7z_RangeDec_IsFinishedOK(&ppmd.rc.dec)) */
+        res = SZ_ERROR_DATA;
+      }
     }
+    if (s.extra)
+      res = (s.res != SZ_OK ? s.res : SZ_ERROR_DATA);
+    else if (s.processed + (size_t)(s.cur - s.begin) != inSize)
+      res = SZ_ERROR_DATA;
   }
   Ppmd7_Free(&ppmd, allocMain);
   return res;
@@ -365,7 +372,9 @@ 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,

libraries/lzma/C/7zStream.c

@@ -1,5 +1,5 @@
 /* 7zStream.c -- 7z Stream functions
-2017-04-03 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -37,7 +37,7 @@ SRes SeqInStream_ReadByte(const ISeqInStream *stream, Byte *buf)
 
 SRes LookInStream_SeekTo(const ILookInStream *stream, UInt64 offset)
 {
-  Int64 t = offset;
+  Int64 t = (Int64)offset;
   return ILookInStream_Seek(stream, &t, SZ_SEEK_SET);
 }
 

libraries/lzma/C/7zTypes.h

@@ -1,11 +1,13 @@
 /* 7zTypes.h -- Basic types
-2018-08-04 : Igor Pavlov : Public domain */
+2021-12-25 : Igor Pavlov : Public domain */
 
 #ifndef __7Z_TYPES_H
 #define __7Z_TYPES_H
 
 #ifdef _WIN32
 /* #include <windows.h> */
+#else
+#include <errno.h>
 #endif
 
 #include <stddef.h>
@@ -43,18 +45,116 @@ EXTERN_C_BEGIN
 typedef int SRes;
 
 
+#ifdef _MSC_VER
+  #if _MSC_VER > 1200
+    #define MY_ALIGN(n) __declspec(align(n))
+  #else
+    #define MY_ALIGN(n)
+  #endif
+#else
+  #define MY_ALIGN(n) __attribute__ ((aligned(n)))
+#endif
+
+
 #ifdef _WIN32
 
 /* typedef DWORD WRes; */
 typedef unsigned WRes;
 #define MY_SRes_HRESULT_FROM_WRes(x) HRESULT_FROM_WIN32(x)
 
-#else
+// #define MY_HRES_ERROR__INTERNAL_ERROR  MY_SRes_HRESULT_FROM_WRes(ERROR_INTERNAL_ERROR)
 
+#else // _WIN32
+
+// #define ENV_HAVE_LSTAT
 typedef int WRes;
-#define MY__FACILITY_WIN32 7
+
-#define MY__FACILITY__WRes MY__FACILITY_WIN32
+// (FACILITY_ERRNO = 0x800) is 7zip's FACILITY constant to represent (errno) errors in HRESULT
-#define MY_SRes_HRESULT_FROM_WRes(x) ((HRESULT)(x) <= 0 ? ((HRESULT)(x)) : ((HRESULT) (((x) & 0x0000FFFF) | (MY__FACILITY__WRes << 16) | 0x80000000)))
+#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)  \
+          | (HRESULT)0x80000000 ))
+
+#define MY_SRes_HRESULT_FROM_WRes(x) \
+  ((HRESULT)(x) <= 0 ? ((HRESULT)(x)) : MY_HRESULT_FROM_errno_CONST_ERROR(x))
+
+// we call macro HRESULT_FROM_WIN32 for system errors (WRes) that are (errno)
+#define HRESULT_FROM_WIN32(x) MY_SRes_HRESULT_FROM_WRes(x)
+
+/*
+#define ERROR_FILE_NOT_FOUND             2L
+#define ERROR_ACCESS_DENIED              5L
+#define ERROR_NO_MORE_FILES              18L
+#define ERROR_LOCK_VIOLATION             33L
+#define ERROR_FILE_EXISTS                80L
+#define ERROR_DISK_FULL                  112L
+#define ERROR_NEGATIVE_SEEK              131L
+#define ERROR_ALREADY_EXISTS             183L
+#define ERROR_DIRECTORY                  267L
+#define ERROR_TOO_MANY_POSTS             298L
+
+#define ERROR_INTERNAL_ERROR             1359L
+#define ERROR_INVALID_REPARSE_DATA       4392L
+#define ERROR_REPARSE_TAG_INVALID        4393L
+#define ERROR_REPARSE_TAG_MISMATCH       4394L
+*/
+
+// we use errno equivalents for some WIN32 errors:
+
+#define ERROR_INVALID_PARAMETER     EINVAL
+#define ERROR_INVALID_FUNCTION      EINVAL
+#define ERROR_ALREADY_EXISTS        EEXIST
+#define ERROR_FILE_EXISTS           EEXIST
+#define ERROR_PATH_NOT_FOUND        ENOENT
+#define ERROR_FILE_NOT_FOUND        ENOENT
+#define ERROR_DISK_FULL             ENOSPC
+// #define ERROR_INVALID_HANDLE        EBADF
+
+// we use FACILITY_WIN32 for errors that has no errno equivalent
+// Too many posts were made to a semaphore.
+#define ERROR_TOO_MANY_POSTS        ((HRESULT)0x8007012AL)
+#define ERROR_INVALID_REPARSE_DATA  ((HRESULT)0x80071128L)
+#define ERROR_REPARSE_TAG_INVALID   ((HRESULT)0x80071129L)
+
+// 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)
+
+/*
+// 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)
+*/
+
+// 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
+#define FILE_ATTRIBUTE_HIDDEN         0x0002
+#define FILE_ATTRIBUTE_SYSTEM         0x0004
+#define FILE_ATTRIBUTE_DIRECTORY      0x0010
+#define FILE_ATTRIBUTE_ARCHIVE        0x0020
+#define FILE_ATTRIBUTE_DEVICE         0x0040
+#define FILE_ATTRIBUTE_NORMAL         0x0080
+#define FILE_ATTRIBUTE_TEMPORARY      0x0100
+#define FILE_ATTRIBUTE_SPARSE_FILE    0x0200
+#define FILE_ATTRIBUTE_REPARSE_POINT  0x0400
+#define FILE_ATTRIBUTE_COMPRESSED     0x0800
+#define FILE_ATTRIBUTE_OFFLINE        0x1000
+#define FILE_ATTRIBUTE_NOT_CONTENT_INDEXED 0x2000
+#define FILE_ATTRIBUTE_ENCRYPTED      0x4000
+
+#define FILE_ATTRIBUTE_UNIX_EXTENSION 0x8000   /* trick for Unix */
 
 #endif
 
@@ -63,6 +163,10 @@ typedef int WRes;
 #define RINOK(x) { int __result__ = (x); if (__result__ != 0) return __result__; }
 #endif
 
+#ifndef RINOK_WRes
+#define RINOK_WRes(x) { WRes __result__ = (x); if (__result__ != 0) return __result__; }
+#endif
+
 typedef unsigned char Byte;
 typedef short Int16;
 typedef unsigned short UInt16;
@@ -75,6 +179,40 @@ typedef int Int32;
 typedef unsigned int UInt32;
 #endif
 
+
+#ifndef _WIN32
+
+typedef int INT;
+typedef Int32 INT32;
+typedef unsigned int UINT;
+typedef UInt32 UINT32;
+typedef INT32 LONG;   // LONG, ULONG and DWORD must be 32-bit for _WIN32 compatibility
+typedef UINT32 ULONG;
+
+#undef DWORD
+typedef UINT32 DWORD;
+
+#define VOID void
+
+#define HRESULT LONG
+
+typedef void *LPVOID;
+// typedef void VOID;
+// typedef ULONG_PTR DWORD_PTR, *PDWORD_PTR;
+// gcc / clang on Unix  : sizeof(long==sizeof(void*) in 32 or 64 bits)
+typedef          long  INT_PTR;
+typedef unsigned long  UINT_PTR;
+typedef          long  LONG_PTR;
+typedef unsigned long  DWORD_PTR;
+
+typedef size_t SIZE_T;
+
+#endif //  _WIN32
+
+
+#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.
@@ -128,25 +266,37 @@ typedef int BoolInt;
 #define MY_CDECL __cdecl
 #define MY_FAST_CALL __fastcall
 
-#else
+#else //  _MSC_VER
 
-#define MY_NO_INLINE
+#if (defined(__GNUC__) && (__GNUC__ >= 4)) \
-#define MY_FORCE_INLINE
+    || (defined(__clang__) && (__clang_major__ >= 4)) \
-#define MY_CDECL
+    || defined(__INTEL_COMPILER) \
-#define MY_FAST_CALL
+    || defined(__xlC__)
-
-/* inline keyword : for C++ / C99 */
-
-/* GCC, clang: */
-/*
-#if defined (__GNUC__) && (__GNUC__ >= 4)
-#define MY_FORCE_INLINE __attribute__((always_inline))
 #define MY_NO_INLINE __attribute__((noinline))
+// #define MY_FORCE_INLINE __attribute__((always_inline)) inline
+#else
+#define MY_NO_INLINE
 #endif
-*/
 
+#define MY_FORCE_INLINE
+
+
+#define MY_CDECL
+
+#if  defined(_M_IX86) \
+  || defined(__i386__)
+// #define MY_FAST_CALL __attribute__((fastcall))
+// #define MY_FAST_CALL __attribute__((cdecl))
+#define MY_FAST_CALL
+#elif defined(MY_CPU_AMD64)
+// #define MY_FAST_CALL __attribute__((ms_abi))
+#define MY_FAST_CALL
+#else
+#define MY_FAST_CALL
 #endif
 
+#endif //  _MSC_VER
+
 
 /* The following interfaces use first parameter as pointer to structure */
 
@@ -335,12 +485,11 @@ struct ISzAlloc
     GCC 4.8.1 : classes with non-public variable members"
 */
 
-#define MY_container_of(ptr, type, m) ((type *)((char *)(1 ? (ptr) : &((type *)0)->m) - MY_offsetof(type, m)))
+#define MY_container_of(ptr, type, m) ((type *)(void *)((char *)(void *)(1 ? (ptr) : &((type *)0)->m) - MY_offsetof(type, m)))
-
 
 #endif
 
-#define CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) ((type *)(ptr))
+#define CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) ((type *)(void *)(ptr))
 
 /*
 #define CONTAINER_FROM_VTBL(ptr, type, m) CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m)
@@ -353,6 +502,7 @@ struct ISzAlloc
 */
 
 
+#define MY_memset_0_ARRAY(a) memset((a), 0, sizeof(a))
 
 #ifdef _WIN32
 

libraries/lzma/C/7zVersion.h

@@ -1,7 +1,7 @@
-#define MY_VER_MAJOR 19
+#define MY_VER_MAJOR 21
-#define MY_VER_MINOR 00
+#define MY_VER_MINOR 07
 #define MY_VER_BUILD 0
-#define MY_VERSION_NUMBERS "19.00"
+#define MY_VERSION_NUMBERS "21.07"
 #define MY_VERSION MY_VERSION_NUMBERS
 
 #ifdef MY_CPU_NAME
@@ -10,12 +10,12 @@
   #define MY_VERSION_CPU MY_VERSION
 #endif
 
-#define MY_DATE "2019-02-21"
+#define MY_DATE "2021-12-26"
 #undef MY_COPYRIGHT
 #undef MY_VERSION_COPYRIGHT_DATE
 #define MY_AUTHOR_NAME "Igor Pavlov"
 #define MY_COPYRIGHT_PD "Igor Pavlov : Public domain"
-#define MY_COPYRIGHT_CR "Copyright (c) 1999-2018 Igor Pavlov"
+#define MY_COPYRIGHT_CR "Copyright (c) 1999-2021 Igor Pavlov"
 
 #ifdef USE_COPYRIGHT_CR
   #define MY_COPYRIGHT MY_COPYRIGHT_CR

libraries/lzma/C/Bcj2.c

@@ -1,5 +1,5 @@
 /* Bcj2.c -- BCJ2 Decoder (Converter for x86 code)
-2018-04-28 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -123,7 +123,7 @@ SRes Bcj2Dec_Decode(CBcj2Dec *p)
         const Byte *src = p->bufs[BCJ2_STREAM_MAIN];
         const Byte *srcLim;
         Byte *dest;
-        SizeT num = p->lims[BCJ2_STREAM_MAIN] - src;
+        SizeT num = (SizeT)(p->lims[BCJ2_STREAM_MAIN] - src);
         
         if (num == 0)
         {
@@ -134,7 +134,7 @@ SRes Bcj2Dec_Decode(CBcj2Dec *p)
         dest = p->dest;
         if (num > (SizeT)(p->destLim - dest))
         {
-          num = p->destLim - dest;
+          num = (SizeT)(p->destLim - dest);
           if (num == 0)
           {
             p->state = BCJ2_DEC_STATE_ORIG;
@@ -168,7 +168,7 @@ SRes Bcj2Dec_Decode(CBcj2Dec *p)
           break;
         }
         
-        num = src - p->bufs[BCJ2_STREAM_MAIN];
+        num = (SizeT)(src - p->bufs[BCJ2_STREAM_MAIN]);
         
         if (src == srcLim)
         {
@@ -228,7 +228,7 @@ SRes Bcj2Dec_Decode(CBcj2Dec *p)
       p->ip += 4;
       val -= p->ip;
       dest = p->dest;
-      rem = p->destLim - dest;
+      rem = (SizeT)(p->destLim - dest);
       
       if (rem < 4)
       {

libraries/lzma/C/Bra.c

@@ -1,5 +1,5 @@
 /* Bra.c -- Converters for RISC code
-2017-04-04 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -22,7 +22,7 @@ SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
     for (;;)
     {
       if (p >= lim)
-        return p - data;
+        return (SizeT)(p - data);
       p += 4;
       if (p[-1] == 0xEB)
         break;
@@ -43,7 +43,7 @@ SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
     for (;;)
     {
       if (p >= lim)
-        return p - data;
+        return (SizeT)(p - data);
       p += 4;
       if (p[-1] == 0xEB)
         break;
@@ -78,7 +78,7 @@ SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
     {
       UInt32 b3;
       if (p > lim)
-        return p - data;
+        return (SizeT)(p - data);
       b1 = p[1];
       b3 = p[3];
       p += 2;
@@ -113,7 +113,7 @@ SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
     {
       UInt32 b3;
       if (p > lim)
-        return p - data;
+        return (SizeT)(p - data);
       b1 = p[1];
       b3 = p[3];
       p += 2;
@@ -162,7 +162,7 @@ SizeT PPC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
     for (;;)
     {
       if (p >= lim)
-        return p - data;
+        return (SizeT)(p - data);
       p += 4;
       /* if ((v & 0xFC000003) == 0x48000001) */
       if ((p[-4] & 0xFC) == 0x48 && (p[-1] & 3) == 1)
@@ -196,7 +196,7 @@ SizeT SPARC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
     for (;;)
     {
       if (p >= lim)
-        return p - data;
+        return (SizeT)(p - data);
       /*
       v = GetBe32(p);
       p += 4;

libraries/lzma/C/Bra86.c

@@ -1,5 +1,5 @@
 /* Bra86.c -- Converter for x86 code (BCJ)
-2017-04-03 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -25,7 +25,7 @@ SizeT x86_Convert(Byte *data, SizeT size, UInt32 ip, UInt32 *state, int encoding
         break;
 
     {
-      SizeT d = (SizeT)(p - data - pos);
+      SizeT d = (SizeT)(p - data) - pos;
       pos = (SizeT)(p - data);
       if (p >= limit)
       {

libraries/lzma/C/Compiler.h

@@ -1,9 +1,13 @@
 /* Compiler.h
-2017-04-03 : Igor Pavlov : Public domain */
+2021-01-05 : Igor Pavlov : Public domain */
 
 #ifndef __7Z_COMPILER_H
 #define __7Z_COMPILER_H
 
+  #ifdef __clang__
+    #pragma clang diagnostic ignored "-Wunused-private-field"
+  #endif
+
 #ifdef _MSC_VER
 
   #ifdef UNDER_CE
@@ -25,6 +29,12 @@
     #pragma warning(disable : 4786) // identifier was truncated to '255' characters in the debug information
   #endif
 
+  #ifdef __clang__
+    #pragma clang diagnostic ignored "-Wdeprecated-declarations"
+    #pragma clang diagnostic ignored "-Wmicrosoft-exception-spec"
+    // #pragma clang diagnostic ignored "-Wreserved-id-macro"
+  #endif
+
 #endif
 
 #define UNUSED_VAR(x) (void)x;

libraries/lzma/C/CpuArch.c

@@ -1,5 +1,5 @@
 /* CpuArch.c -- CPU specific code
-2018-02-18: Igor Pavlov : Public domain */
+2021-07-13 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -55,6 +55,47 @@ static UInt32 CheckFlag(UInt32 flag)
 #define CHECK_CPUID_IS_SUPPORTED
 #endif
 
+#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
@@ -99,18 +140,20 @@ void MyCPUID(UInt32 function, UInt32 *a, UInt32 *b, UInt32 *c, UInt32 *d)
   #endif
       "=c" (*c) ,
       "=d" (*d)
-    : "0" (function)) ;
+    : "0" (function), "c"(0) ) ;
 
   #endif
   
   #else
 
   int CPUInfo[4];
-  __cpuid(CPUInfo, function);
+
-  *a = CPUInfo[0];
+  MY__cpuidex(CPUInfo, (int)function, 0);
-  *b = CPUInfo[1];
+
-  *c = CPUInfo[2];
+  *a = (UInt32)CPUInfo[0];
-  *d = CPUInfo[3];
+  *b = (UInt32)CPUInfo[1];
+  *c = (UInt32)CPUInfo[2];
+  *d = (UInt32)CPUInfo[3];
 
   #endif
 }
@@ -174,7 +217,7 @@ BoolInt CPU_Is_InOrder()
 }
 
 #if !defined(MY_CPU_AMD64) && defined(_WIN32)
-#include <windows.h>
+#include <Windows.h>
 static BoolInt CPU_Sys_Is_SSE_Supported()
 {
   OSVERSIONINFO vi;
@@ -188,13 +231,101 @@ static BoolInt CPU_Sys_Is_SSE_Supported()
 #define CHECK_SYS_SSE_SUPPORT
 #endif
 
-BoolInt CPU_Is_Aes_Supported()
+
+static UInt32 X86_CPUID_ECX_Get_Flags()
+{
+  Cx86cpuid p;
+  CHECK_SYS_SSE_SUPPORT
+  if (!x86cpuid_CheckAndRead(&p))
+    return 0;
+  return p.c;
+}
+
+BoolInt CPU_IsSupported_AES()
+{
+  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;
   CHECK_SYS_SSE_SUPPORT
   if (!x86cpuid_CheckAndRead(&p))
     return False;
-  return (p.c >> 25) & 1;
+
+  if (p.maxFunc < 7)
+    return False;
+  {
+    UInt32 d[4] = { 0 };
+    MyCPUID(7, &d[0], &d[1], &d[2], &d[3]);
+    return (d[1] >> 29) & 1;
+  }
+}
+
+// #include <stdio.h>
+
+#ifdef _WIN32
+#include <Windows.h>
+#endif
+
+BoolInt CPU_IsSupported_AVX2()
+{
+  Cx86cpuid p;
+  CHECK_SYS_SSE_SUPPORT
+
+  #ifdef _WIN32
+  #define MY__PF_XSAVE_ENABLED  17
+  if (!IsProcessorFeaturePresent(MY__PF_XSAVE_ENABLED))
+    return False;
+  #endif
+
+  if (!x86cpuid_CheckAndRead(&p))
+    return False;
+  if (p.maxFunc < 7)
+    return False;
+  {
+    UInt32 d[4] = { 0 };
+    MyCPUID(7, &d[0], &d[1], &d[2], &d[3]);
+    // printf("\ncpuid(7): ebx=%8x ecx=%8x\n", d[1], d[2]);
+    return 1
+      & (d[1] >> 5); // avx2
+  }
+}
+
+BoolInt CPU_IsSupported_VAES_AVX2()
+{
+  Cx86cpuid p;
+  CHECK_SYS_SSE_SUPPORT
+
+  #ifdef _WIN32
+  #define MY__PF_XSAVE_ENABLED  17
+  if (!IsProcessorFeaturePresent(MY__PF_XSAVE_ENABLED))
+    return False;
+  #endif
+
+  if (!x86cpuid_CheckAndRead(&p))
+    return False;
+  if (p.maxFunc < 7)
+    return False;
+  {
+    UInt32 d[4] = { 0 };
+    MyCPUID(7, &d[0], &d[1], &d[2], &d[3]);
+    // printf("\ncpuid(7): ebx=%8x ecx=%8x\n", d[1], d[2]);
+    return 1
+      & (d[1] >> 5) // avx2
+      // & (d[1] >> 31) // avx512vl
+      & (d[2] >> 9); // vaes // VEX-256/EVEX
+  }
 }
 
 BoolInt CPU_IsSupported_PageGB()
@@ -215,4 +346,133 @@ BoolInt CPU_IsSupported_PageGB()
   }
 }
 
+
+#elif defined(MY_CPU_ARM_OR_ARM64)
+
+#ifdef _WIN32
+
+#include <Windows.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; }
+
+#else
+
+#if defined(__APPLE__)
+
+/*
+#include <stdio.h>
+#include <string.h>
+static void Print_sysctlbyname(const char *name)
+{
+  size_t bufSize = 256;
+  char buf[256];
+  int res = sysctlbyname(name, &buf, &bufSize, NULL, 0);
+  {
+    int i;
+    printf("\nres = %d : %s : '%s' : bufSize = %d, numeric", res, name, buf, (unsigned)bufSize);
+    for (i = 0; i < 20; i++)
+      printf(" %2x", (unsigned)(Byte)buf[i]);
+
+  }
+}
+*/
+
+static BoolInt My_sysctlbyname_Get_BoolInt(const char *name)
+{
+  UInt32 val = 0;
+  if (My_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");
+}
+
+BoolInt CPU_IsSupported_NEON(void)
+{
+  return My_sysctlbyname_Get_BoolInt("hw.optional.neon");
+}
+
+#ifdef MY_CPU_ARM64
+#define APPLE_CRYPTO_SUPPORT_VAL 1
+#else
+#define APPLE_CRYPTO_SUPPORT_VAL 0
+#endif
+
+BoolInt CPU_IsSupported_SHA1(void) { return APPLE_CRYPTO_SUPPORT_VAL; }
+BoolInt CPU_IsSupported_SHA2(void) { return APPLE_CRYPTO_SUPPORT_VAL; }
+BoolInt CPU_IsSupported_AES (void) { return APPLE_CRYPTO_SUPPORT_VAL; }
+
+
+#else // __APPLE__
+
+#include <sys/auxv.h>
+
+#define USE_HWCAP
+
+#ifdef USE_HWCAP
+
+#include <asm/hwcap.h>
+
+  #define MY_HWCAP_CHECK_FUNC_2(name1, name2) \
+  BoolInt CPU_IsSupported_ ## name1() { return (getauxval(AT_HWCAP)  & (HWCAP_  ## name2)) ? 1 : 0; }
+
+#ifdef MY_CPU_ARM64
+  #define MY_HWCAP_CHECK_FUNC(name) \
+  MY_HWCAP_CHECK_FUNC_2(name, name)
+  MY_HWCAP_CHECK_FUNC_2(NEON, ASIMD)
+// MY_HWCAP_CHECK_FUNC (ASIMD)
+#elif defined(MY_CPU_ARM)
+  #define MY_HWCAP_CHECK_FUNC(name) \
+  BoolInt CPU_IsSupported_ ## name() { return (getauxval(AT_HWCAP2) & (HWCAP2_ ## name)) ? 1 : 0; }
+  MY_HWCAP_CHECK_FUNC_2(NEON, NEON)
+#endif
+
+#else // USE_HWCAP
+
+  #define MY_HWCAP_CHECK_FUNC(name) \
+  BoolInt CPU_IsSupported_ ## name() { return 0; }
+  MY_HWCAP_CHECK_FUNC(NEON)
+
+#endif // USE_HWCAP
+
+MY_HWCAP_CHECK_FUNC (CRC32)
+MY_HWCAP_CHECK_FUNC (SHA1)
+MY_HWCAP_CHECK_FUNC (SHA2)
+MY_HWCAP_CHECK_FUNC (AES)
+
+#endif // __APPLE__
+#endif // _WIN32
+
+#endif // MY_CPU_ARM_OR_ARM64
+
+
+
+#ifdef __APPLE__
+
+#include <sys/sysctl.h>
+
+int My_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)
+{
+  size_t bufSize = sizeof(*val);
+  int res = My_sysctlbyname_Get(name, val, &bufSize);
+  if (res == 0 && bufSize != sizeof(*val))
+    return EFAULT;
+  return res;
+}
+
 #endif

libraries/lzma/C/CpuArch.h

@@ -1,5 +1,5 @@
 /* CpuArch.h -- CPU specific code
-2018-02-18 : Igor Pavlov : Public domain */
+2021-07-13 : Igor Pavlov : Public domain */
 
 #ifndef __CPU_ARCH_H
 #define __CPU_ARCH_H
@@ -14,6 +14,10 @@ MY_CPU_BE means that CPU is BIG ENDIAN.
 If MY_CPU_LE and MY_CPU_BE are not defined, we don't know about ENDIANNESS of platform.
 
 MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned memory accesses.
+
+MY_CPU_64BIT means that processor can work with 64-bit registers.
+  MY_CPU_64BIT can be used to select fast code branch
+  MY_CPU_64BIT doesn't mean that (sizeof(void *) == 8)
 */
 
 #if  defined(_M_X64) \
@@ -24,8 +28,10 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
   #define MY_CPU_AMD64
   #ifdef __ILP32__
     #define MY_CPU_NAME "x32"
+    #define MY_CPU_SIZEOF_POINTER 4
   #else
     #define MY_CPU_NAME "x64"
+    #define MY_CPU_SIZEOF_POINTER 8
   #endif
   #define MY_CPU_64BIT
 #endif
@@ -35,7 +41,8 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
   || defined(__i386__)
   #define MY_CPU_X86
   #define MY_CPU_NAME "x86"
-  #define MY_CPU_32BIT
+  /* #define MY_CPU_32BIT */
+  #define MY_CPU_SIZEOF_POINTER 4
 #endif
 
 
@@ -59,8 +66,14 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
   || defined(__THUMBEL__) \
   || defined(__THUMBEB__)
   #define MY_CPU_ARM
-  #define MY_CPU_NAME "arm"
+
-  #define MY_CPU_32BIT
+  #if defined(__thumb__) || defined(__THUMBEL__) || defined(_M_ARMT)
+    #define MY_CPU_NAME "armt"
+  #else
+    #define MY_CPU_NAME "arm"
+  #endif
+  /* #define MY_CPU_32BIT */
+  #define MY_CPU_SIZEOF_POINTER 4
 #endif
 
 
@@ -84,17 +97,29 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
 
 
 #if  defined(__ppc64__) \
-  || defined(__powerpc64__)
+  || defined(__powerpc64__) \
+  || defined(__ppc__) \
+  || defined(__powerpc__) \
+  || defined(__PPC__) \
+  || defined(_POWER)
+
+#if  defined(__ppc64__) \
+  || defined(__powerpc64__) \
+  || defined(_LP64) \
+  || defined(__64BIT__)
   #ifdef __ILP32__
     #define MY_CPU_NAME "ppc64-32"
+    #define MY_CPU_SIZEOF_POINTER 4
   #else
     #define MY_CPU_NAME "ppc64"
+    #define MY_CPU_SIZEOF_POINTER 8
   #endif
   #define MY_CPU_64BIT
-#elif defined(__ppc__) \
+#else
-  || defined(__powerpc__)
   #define MY_CPU_NAME "ppc"
-  #define MY_CPU_32BIT
+  #define MY_CPU_SIZEOF_POINTER 4
+  /* #define MY_CPU_32BIT */
+#endif
 #endif
 
 
@@ -111,6 +136,10 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
 #define MY_CPU_X86_OR_AMD64
 #endif
 
+#if defined(MY_CPU_ARM) || defined(MY_CPU_ARM64)
+#define MY_CPU_ARM_OR_ARM64
+#endif
+
 
 #ifdef _WIN32
 
@@ -170,6 +199,40 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
   #error Stop_Compiling_Bad_32_64_BIT
 #endif
 
+#ifdef __SIZEOF_POINTER__
+  #ifdef MY_CPU_SIZEOF_POINTER
+    #if MY_CPU_SIZEOF_POINTER != __SIZEOF_POINTER__
+      #error Stop_Compiling_Bad_MY_CPU_PTR_SIZE
+    #endif
+  #else
+    #define MY_CPU_SIZEOF_POINTER  __SIZEOF_POINTER__
+  #endif
+#endif
+
+#if defined(MY_CPU_SIZEOF_POINTER) && (MY_CPU_SIZEOF_POINTER == 4)
+#if defined (_LP64)
+      #error Stop_Compiling_Bad_MY_CPU_PTR_SIZE
+#endif
+#endif
+
+#ifdef _MSC_VER
+  #if _MSC_VER >= 1300
+    #define MY_CPU_pragma_pack_push_1   __pragma(pack(push, 1))
+    #define MY_CPU_pragma_pop           __pragma(pack(pop))
+  #else
+    #define MY_CPU_pragma_pack_push_1
+    #define MY_CPU_pragma_pop
+  #endif
+#else
+  #ifdef __xlC__
+    #define MY_CPU_pragma_pack_push_1   _Pragma("pack(1)")
+    #define MY_CPU_pragma_pop           _Pragma("pack()")
+  #else
+    #define MY_CPU_pragma_pack_push_1   _Pragma("pack(push, 1)")
+    #define MY_CPU_pragma_pop           _Pragma("pack(pop)")
+  #endif
+#endif
+
 
 #ifndef MY_CPU_NAME
   #ifdef MY_CPU_LE
@@ -189,8 +252,12 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
 
 #ifdef MY_CPU_LE
   #if defined(MY_CPU_X86_OR_AMD64) \
-      || defined(MY_CPU_ARM64) \
+      || defined(MY_CPU_ARM64)
-      || defined(__ARM_FEATURE_UNALIGNED)
+    #define MY_CPU_LE_UNALIGN
+    #define MY_CPU_LE_UNALIGN_64
+  #elif defined(__ARM_FEATURE_UNALIGNED)
+    /* gcc9 for 32-bit arm can use LDRD instruction that requires 32-bit alignment.
+       So we can't use unaligned 64-bit operations. */
     #define MY_CPU_LE_UNALIGN
   #endif
 #endif
@@ -200,11 +267,15 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
 
 #define GetUi16(p) (*(const UInt16 *)(const void *)(p))
 #define GetUi32(p) (*(const UInt32 *)(const void *)(p))
+#ifdef MY_CPU_LE_UNALIGN_64
 #define GetUi64(p) (*(const UInt64 *)(const void *)(p))
+#endif
 
-#define SetUi16(p, v) { *(UInt16 *)(p) = (v); }
+#define SetUi16(p, v) { *(UInt16 *)(void *)(p) = (v); }
-#define SetUi32(p, v) { *(UInt32 *)(p) = (v); }
+#define SetUi32(p, v) { *(UInt32 *)(void *)(p) = (v); }
-#define SetUi64(p, v) { *(UInt64 *)(p) = (v); }
+#ifdef MY_CPU_LE_UNALIGN_64
+#define SetUi64(p, v) { *(UInt64 *)(void *)(p) = (v); }
+#endif
 
 #else
 
@@ -218,8 +289,6 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
     ((UInt32)((const Byte *)(p))[2] << 16) | \
     ((UInt32)((const Byte *)(p))[3] << 24))
 
-#define GetUi64(p) (GetUi32(p) | ((UInt64)GetUi32(((const Byte *)(p)) + 4) << 32))
-
 #define SetUi16(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \
     _ppp_[0] = (Byte)_vvv_; \
     _ppp_[1] = (Byte)(_vvv_ >> 8); }
@@ -230,19 +299,29 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
     _ppp_[2] = (Byte)(_vvv_ >> 16); \
     _ppp_[3] = (Byte)(_vvv_ >> 24); }
 
+#endif
+
+
+#ifndef MY_CPU_LE_UNALIGN_64
+
+#define GetUi64(p) (GetUi32(p) | ((UInt64)GetUi32(((const Byte *)(p)) + 4) << 32))
+
 #define SetUi64(p, v) { Byte *_ppp2_ = (Byte *)(p); UInt64 _vvv2_ = (v); \
     SetUi32(_ppp2_    , (UInt32)_vvv2_); \
     SetUi32(_ppp2_ + 4, (UInt32)(_vvv2_ >> 32)); }
 
 #endif
 
+
+
+
 #ifdef __has_builtin
   #define MY__has_builtin(x) __has_builtin(x)
 #else
   #define MY__has_builtin(x) 0
 #endif
 
-#if defined(MY_CPU_LE_UNALIGN) && /* defined(_WIN64) && */ (_MSC_VER >= 1300)
+#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 */
 
@@ -253,8 +332,8 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
 #pragma intrinsic(_byteswap_uint64)
 
 /* #define GetBe16(p) _byteswap_ushort(*(const UInt16 *)(const Byte *)(p)) */
-#define GetBe32(p) _byteswap_ulong(*(const UInt32 *)(const Byte *)(p))
+#define GetBe32(p) _byteswap_ulong (*(const UInt32 *)(const void *)(p))
-#define GetBe64(p) _byteswap_uint64(*(const UInt64 *)(const Byte *)(p))
+#define GetBe64(p) _byteswap_uint64(*(const UInt64 *)(const void *)(p))
 
 #define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = _byteswap_ulong(v)
 
@@ -262,9 +341,9 @@ MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned mem
        (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3))) \
     || (defined(__clang__) && MY__has_builtin(__builtin_bswap16)) )
 
-/* #define GetBe16(p) __builtin_bswap16(*(const UInt16 *)(const Byte *)(p)) */
+/* #define GetBe16(p) __builtin_bswap16(*(const UInt16 *)(const void *)(p)) */
-#define GetBe32(p) __builtin_bswap32(*(const UInt32 *)(const Byte *)(p))
+#define GetBe32(p) __builtin_bswap32(*(const UInt32 *)(const void *)(p))
-#define GetBe64(p) __builtin_bswap64(*(const UInt64 *)(const Byte *)(p))
+#define GetBe64(p) __builtin_bswap64(*(const UInt64 *)(const void *)(p))
 
 #define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = __builtin_bswap32(v)
 
@@ -325,10 +404,37 @@ int x86cpuid_GetFirm(const Cx86cpuid *p);
 #define x86cpuid_GetModel(ver)  (((ver >> 12) &  0xF0) | ((ver >> 4) & 0xF))
 #define x86cpuid_GetStepping(ver) (ver & 0xF)
 
-BoolInt CPU_Is_InOrder();
+BoolInt CPU_Is_InOrder(void);
-BoolInt CPU_Is_Aes_Supported();
-BoolInt CPU_IsSupported_PageGB();
 
+BoolInt CPU_IsSupported_AES(void);
+BoolInt CPU_IsSupported_AVX2(void);
+BoolInt CPU_IsSupported_VAES_AVX2(void);
+BoolInt CPU_IsSupported_SSSE3(void);
+BoolInt CPU_IsSupported_SSE41(void);
+BoolInt CPU_IsSupported_SHA(void);
+BoolInt CPU_IsSupported_PageGB(void);
+
+#elif defined(MY_CPU_ARM_OR_ARM64)
+
+BoolInt CPU_IsSupported_CRC32(void);
+BoolInt CPU_IsSupported_NEON(void);
+
+#if defined(_WIN32)
+BoolInt CPU_IsSupported_CRYPTO(void);
+#define CPU_IsSupported_SHA1  CPU_IsSupported_CRYPTO
+#define CPU_IsSupported_SHA2  CPU_IsSupported_CRYPTO
+#define CPU_IsSupported_AES   CPU_IsSupported_CRYPTO
+#else
+BoolInt CPU_IsSupported_SHA1(void);
+BoolInt CPU_IsSupported_SHA2(void);
+BoolInt CPU_IsSupported_AES(void);
+#endif
+
+#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);
 #endif
 
 EXTERN_C_END

libraries/lzma/C/Delta.c

@@ -1,5 +1,5 @@
 /* Delta.c -- Delta converter
-2009-05-26 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -12,53 +12,158 @@ void Delta_Init(Byte *state)
     state[i] = 0;
 }
 
-static void MyMemCpy(Byte *dest, const Byte *src, unsigned size)
-{
-  unsigned i;
-  for (i = 0; i < size; i++)
-    dest[i] = src[i];
-}
 
 void Delta_Encode(Byte *state, unsigned delta, Byte *data, SizeT size)
 {
-  Byte buf[DELTA_STATE_SIZE];
+  Byte temp[DELTA_STATE_SIZE];
-  unsigned j = 0;
+
-  MyMemCpy(buf, state, delta);
+  if (size == 0)
+    return;
+
   {
-    SizeT i;
+    unsigned i = 0;
-    for (i = 0; i < size;)
+    do
+      temp[i] = state[i];
+    while (++i != delta);
+  }
+
+  if (size <= delta)
+  {
+    unsigned i = 0, k;
+    do
     {
-      for (j = 0; j < delta && i < size; i++, j++)
+      Byte b = *data;
+      *data++ = (Byte)(b - temp[i]);
+      temp[i] = b;
+    }
+    while (++i != size);
+    
+    k = 0;
+    
+    do
+    {
+      if (i == delta)
+        i = 0;
+      state[k] = temp[i++];
+    }
+    while (++k != delta);
+    
+    return;
+  }
+    
+  {
+    Byte *p = data + size - delta;
+    {
+      unsigned i = 0;
+      do
+        state[i] = *p++;
+      while (++i != delta);
+    }
+    {
+      const Byte *lim = data + delta;
+      ptrdiff_t dif = -(ptrdiff_t)delta;
+      
+      if (((ptrdiff_t)size + dif) & 1)
       {
-        Byte b = data[i];
+        --p;  *p = (Byte)(*p - p[dif]);
-        data[i] = (Byte)(b - buf[j]);
-        buf[j] = b;
       }
+
+      while (p != lim)
+      {
+        --p;  *p = (Byte)(*p - p[dif]);
+        --p;  *p = (Byte)(*p - p[dif]);
+      }
+      
+      dif = -dif;
+      
+      do
+      {
+        --p;  *p = (Byte)(*p - temp[--dif]);
+      }
+      while (dif != 0);
     }
   }
-  if (j == delta)
-    j = 0;
-  MyMemCpy(state, buf + j, delta - j);
-  MyMemCpy(state + delta - j, buf, j);
 }
 
+
 void Delta_Decode(Byte *state, unsigned delta, Byte *data, SizeT size)
 {
-  Byte buf[DELTA_STATE_SIZE];
+  unsigned i;
-  unsigned j = 0;
+  const Byte *lim;
-  MyMemCpy(buf, state, delta);
+
+  if (size == 0)
+    return;
+  
+  i = 0;
+  lim = data + size;
+  
+  if (size <= delta)
   {
-    SizeT i;
+    do
-    for (i = 0; i < size;)
+      *data = (Byte)(*data + state[i++]);
+    while (++data != lim);
+
+    for (; delta != i; state++, delta--)
+      *state = state[i];
+    data -= i;
+  }
+  else
+  {
+    /*
+    #define B(n) b ## n
+    #define I(n) Byte B(n) = state[n];
+    #define U(n) { B(n) = (Byte)((B(n)) + *data++); data[-1] = (B(n)); }
+    #define F(n) if (data != lim) { U(n) }
+
+    if (delta == 1)
     {
-      for (j = 0; j < delta && i < size; i++, j++)
+      I(0)
+      if ((lim - data) & 1) { U(0) }
+      while (data != lim) { U(0) U(0) }
+      data -= 1;
+    }
+    else if (delta == 2)
+    {
+      I(0) I(1)
+      lim -= 1; while (data < lim) { U(0) U(1) }
+      lim += 1; F(0)
+      data -= 2;
+    }
+    else if (delta == 3)
+    {
+      I(0) I(1) I(2)
+      lim -= 2; while (data < lim) { U(0) U(1) U(2) }
+      lim += 2; F(0) F(1)
+      data -= 3;
+    }
+    else if (delta == 4)
+    {
+      I(0) I(1) I(2) I(3)
+      lim -= 3; while (data < lim) { U(0) U(1) U(2) U(3) }
+      lim += 3; F(0) F(1) F(2)
+      data -= 4;
+    }
+    else
+    */
+    {
+      do
       {
-        buf[j] = data[i] = (Byte)(buf[j] + data[i]);
+        *data = (Byte)(*data + state[i++]);
+        data++;
+      }
+      while (i != delta);
+  
+      {
+        ptrdiff_t dif = -(ptrdiff_t)delta;
+        do
+          *data = (Byte)(*data + data[dif]);
+        while (++data != lim);
+        data += dif;
       }
     }
   }
-  if (j == delta)
+
-    j = 0;
+  do
-  MyMemCpy(state, buf + j, delta - j);
+    *state++ = *data;
-  MyMemCpy(state + delta - j, buf, j);
+  while (++data != lim);
 }

libraries/lzma/C/LzFind.h

@@ -1,5 +1,5 @@
 /* LzFind.h -- Match finder for LZ algorithms
-2017-06-10 : Igor Pavlov : Public domain */
+2021-07-13 : Igor Pavlov : Public domain */
 
 #ifndef __LZ_FIND_H
 #define __LZ_FIND_H
@@ -15,7 +15,7 @@ typedef struct _CMatchFinder
   Byte *buffer;
   UInt32 pos;
   UInt32 posLimit;
-  UInt32 streamPos;
+  UInt32 streamPos;  /* wrap over Zero is allowed (streamPos < pos). Use (UInt32)(streamPos - pos) */
   UInt32 lenLimit;
 
   UInt32 cyclicBufferPos;
@@ -51,17 +51,19 @@ typedef struct _CMatchFinder
   UInt64 expectedDataSize;
 } CMatchFinder;
 
-#define Inline_MatchFinder_GetPointerToCurrentPos(p) ((p)->buffer)
+#define Inline_MatchFinder_GetPointerToCurrentPos(p) ((const Byte *)(p)->buffer)
 
-#define Inline_MatchFinder_GetNumAvailableBytes(p) ((p)->streamPos - (p)->pos)
+#define Inline_MatchFinder_GetNumAvailableBytes(p) ((UInt32)((p)->streamPos - (p)->pos))
 
+/*
 #define Inline_MatchFinder_IsFinishedOK(p) \
     ((p)->streamEndWasReached \
         && (p)->streamPos == (p)->pos \
         && (!(p)->directInput || (p)->directInputRem == 0))
+*/
       
 int MatchFinder_NeedMove(CMatchFinder *p);
-Byte *MatchFinder_GetPointerToCurrentPos(CMatchFinder *p);
+/* Byte *MatchFinder_GetPointerToCurrentPos(CMatchFinder *p); */
 void MatchFinder_MoveBlock(CMatchFinder *p);
 void MatchFinder_ReadIfRequired(CMatchFinder *p);
 
@@ -76,10 +78,21 @@ int MatchFinder_Create(CMatchFinder *p, UInt32 historySize,
     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);
+// void MatchFinder_ReduceOffsets(CMatchFinder *p, UInt32 subValue);
+
+/*
+#define Inline_MatchFinder_InitPos(p, val) \
+    (p)->pos = (val); \
+    (p)->streamPos = (val);
+*/
+
+#define Inline_MatchFinder_ReduceOffsets(p, subValue) \
+    (p)->pos -= (subValue); \
+    (p)->streamPos -= (subValue);
+
 
 UInt32 * GetMatchesSpec1(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const Byte *buffer, CLzRef *son,
-    UInt32 _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 _cutValue,
+    size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 _cutValue,
     UInt32 *distances, UInt32 maxLen);
 
 /*
@@ -91,7 +104,7 @@ Conditions:
 typedef void (*Mf_Init_Func)(void *object);
 typedef UInt32 (*Mf_GetNumAvailableBytes_Func)(void *object);
 typedef const Byte * (*Mf_GetPointerToCurrentPos_Func)(void *object);
-typedef UInt32 (*Mf_GetMatches_Func)(void *object, UInt32 *distances);
+typedef UInt32 * (*Mf_GetMatches_Func)(void *object, UInt32 *distances);
 typedef void (*Mf_Skip_Func)(void *object, UInt32);
 
 typedef struct _IMatchFinder
@@ -101,21 +114,23 @@ typedef struct _IMatchFinder
   Mf_GetPointerToCurrentPos_Func GetPointerToCurrentPos;
   Mf_GetMatches_Func GetMatches;
   Mf_Skip_Func Skip;
-} IMatchFinder;
+} IMatchFinder2;
 
-void MatchFinder_CreateVTable(CMatchFinder *p, IMatchFinder *vTable);
+void MatchFinder_CreateVTable(CMatchFinder *p, IMatchFinder2 *vTable);
 
 void MatchFinder_Init_LowHash(CMatchFinder *p);
 void MatchFinder_Init_HighHash(CMatchFinder *p);
-void MatchFinder_Init_3(CMatchFinder *p, int readData);
+void MatchFinder_Init_4(CMatchFinder *p);
 void MatchFinder_Init(CMatchFinder *p);
 
-UInt32 Bt3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances);
+UInt32* Bt3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances);
-UInt32 Hc3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances);
+UInt32* Hc3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances);
 
 void Bt3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num);
 void Hc3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num);
 
+void LzFindPrepare(void);
+
 EXTERN_C_END
 
 #endif

libraries/lzma/C/LzFindMt.h

@@ -1,5 +1,5 @@
 /* LzFindMt.h -- multithreaded Match finder for LZ algorithms
-2018-07-04 : Igor Pavlov : Public domain */
+2021-07-12 : Igor Pavlov : Public domain */
 
 #ifndef __LZ_FIND_MT_H
 #define __LZ_FIND_MT_H
@@ -9,31 +9,26 @@
 
 EXTERN_C_BEGIN
 
-#define kMtHashBlockSize (1 << 13)
-#define kMtHashNumBlocks (1 << 3)
-#define kMtHashNumBlocksMask (kMtHashNumBlocks - 1)
-
-#define kMtBtBlockSize (1 << 14)
-#define kMtBtNumBlocks (1 << 6)
-#define kMtBtNumBlocksMask (kMtBtNumBlocks - 1)
-
 typedef struct _CMtSync
 {
+  UInt32 numProcessedBlocks;
+  CThread thread;
+  UInt64 affinity;
+
   BoolInt wasCreated;
   BoolInt needStart;
+  BoolInt csWasInitialized;
+  BoolInt csWasEntered;
+
   BoolInt exit;
   BoolInt stopWriting;
 
-  CThread thread;
   CAutoResetEvent canStart;
-  CAutoResetEvent wasStarted;
   CAutoResetEvent wasStopped;
   CSemaphore freeSemaphore;
   CSemaphore filledSemaphore;
-  BoolInt csWasInitialized;
-  BoolInt csWasEntered;
   CCriticalSection cs;
-  UInt32 numProcessedBlocks;
+  // UInt32 numBlocks_Sent;
 } CMtSync;
 
 typedef UInt32 * (*Mf_Mix_Matches)(void *p, UInt32 matchMinPos, UInt32 *distances);
@@ -49,18 +44,23 @@ typedef struct _CMatchFinderMt
   /* LZ */
   const Byte *pointerToCurPos;
   UInt32 *btBuf;
-  UInt32 btBufPos;
+  const UInt32 *btBufPos;
-  UInt32 btBufPosLimit;
+  const UInt32 *btBufPosLimit;
   UInt32 lzPos;
   UInt32 btNumAvailBytes;
 
   UInt32 *hash;
   UInt32 fixedHashSize;
+  // UInt32 hash4Mask;
   UInt32 historySize;
   const UInt32 *crc;
 
   Mf_Mix_Matches MixMatchesFunc;
-  
+  UInt32 failure_LZ_BT; // failure in BT transfered to LZ
+  // UInt32 failure_LZ_LZ; // failure in LZ tables
+  UInt32 failureBuf[1];
+  // UInt32 crc[256];
+
   /* LZ + BT */
   CMtSync btSync;
   Byte btDummy[kMtCacheLineDummy];
@@ -70,6 +70,8 @@ typedef struct _CMatchFinderMt
   UInt32 hashBufPos;
   UInt32 hashBufPosLimit;
   UInt32 hashNumAvail;
+  UInt32 failure_BT;
+
 
   CLzRef *son;
   UInt32 matchMaxLen;
@@ -77,7 +79,7 @@ typedef struct _CMatchFinderMt
   UInt32 pos;
   const Byte *buffer;
   UInt32 cyclicBufferPos;
-  UInt32 cyclicBufferSize; /* it must be historySize + 1 */
+  UInt32 cyclicBufferSize; /* it must be = (historySize + 1) */
   UInt32 cutValue;
 
   /* BT + Hash */
@@ -87,13 +89,19 @@ typedef struct _CMatchFinderMt
   /* Hash */
   Mf_GetHeads GetHeadsFunc;
   CMatchFinder *MatchFinder;
+  // CMatchFinder MatchFinder;
 } CMatchFinderMt;
 
+// only for Mt part
 void MatchFinderMt_Construct(CMatchFinderMt *p);
 void MatchFinderMt_Destruct(CMatchFinderMt *p, ISzAllocPtr alloc);
+
 SRes MatchFinderMt_Create(CMatchFinderMt *p, UInt32 historySize, UInt32 keepAddBufferBefore,
     UInt32 matchMaxLen, UInt32 keepAddBufferAfter, ISzAllocPtr alloc);
-void MatchFinderMt_CreateVTable(CMatchFinderMt *p, IMatchFinder *vTable);
+void MatchFinderMt_CreateVTable(CMatchFinderMt *p, IMatchFinder2 *vTable);
+
+/* call MatchFinderMt_InitMt() before IMatchFinder::Init() */
+SRes MatchFinderMt_InitMt(CMatchFinderMt *p);
 void MatchFinderMt_ReleaseStream(CMatchFinderMt *p);
 
 EXTERN_C_END

libraries/lzma/C/LzFindOpt.c

@@ -0,0 +1,578 @@
+/* LzFindOpt.c -- multithreaded Match finder for LZ algorithms
+2021-07-13 : Igor Pavlov : Public domain */
+
+#include "Precomp.h"
+
+#include "CpuArch.h"
+#include "LzFind.h"
+
+// #include "LzFindMt.h"
+
+// #define LOG_ITERS
+
+// #define LOG_THREAD
+
+#ifdef LOG_THREAD
+#include <stdio.h>
+#define PRF(x) x
+#else
+// #define PRF(x)
+#endif
+
+#ifdef LOG_ITERS
+#include <stdio.h>
+UInt64 g_NumIters_Tree;
+UInt64 g_NumIters_Loop;
+UInt64 g_NumIters_Bytes;
+#define LOG_ITER(x) x
+#else
+#define LOG_ITER(x)
+#endif
+
+// ---------- BT THREAD ----------
+
+#define USE_SON_PREFETCH
+#define USE_LONG_MATCH_OPT
+
+#define kEmptyHashValue 0
+
+// #define CYC_TO_POS_OFFSET 0
+
+// #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,
+    UInt32 _cutValue, UInt32 *d, size_t _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size, UInt32 *posRes)
+{
+  do
+  {
+    UInt32 delta;
+    if (hash == size)
+      break;
+    delta = *hash++;
+
+    if (delta == 0 || delta > (UInt32)pos)
+      return NULL;
+
+    lenLimit++;
+
+    if (delta == (UInt32)pos)
+    {
+      CLzRef *ptr1 = son + ((size_t)pos << 1) - CYC_TO_POS_OFFSET * 2;
+      *d++ = 0;
+      ptr1[0] = kEmptyHashValue;
+      ptr1[1] = kEmptyHashValue;
+    }
+else
+{
+  UInt32 *_distances = ++d;
+
+  CLzRef *ptr0 = son + ((size_t)(pos) << 1) - CYC_TO_POS_OFFSET * 2 + 1;
+  CLzRef *ptr1 = son + ((size_t)(pos) << 1) - CYC_TO_POS_OFFSET * 2;
+
+  const Byte *len0 = cur, *len1 = cur;
+  UInt32 cutValue = _cutValue;
+  const Byte *maxLen = cur + _maxLen;
+
+  for (LOG_ITER(g_NumIters_Tree++);;)
+  {
+    LOG_ITER(g_NumIters_Loop++);
+    {
+      const ptrdiff_t diff = (ptrdiff_t)0 - (ptrdiff_t)delta;
+      CLzRef *pair = son + ((size_t)(((ptrdiff_t)pos - CYC_TO_POS_OFFSET) + diff) << 1);
+      const Byte *len = (len0 < len1 ? len0 : len1);
+
+    #ifdef USE_SON_PREFETCH
+      const UInt32 pair0 = *pair;
+    #endif
+
+      if (len[diff] == len[0])
+      {
+        if (++len != lenLimit && len[diff] == len[0])
+          while (++len != lenLimit)
+          {
+            LOG_ITER(g_NumIters_Bytes++);
+            if (len[diff] != len[0])
+              break;
+          }
+        if (maxLen < len)
+        {
+          maxLen = len;
+          *d++ = (UInt32)(len - cur);
+          *d++ = delta - 1;
+          
+          if (len == lenLimit)
+          {
+            const UInt32 pair1 = pair[1];
+            *ptr1 =
+              #ifdef USE_SON_PREFETCH
+                pair0;
+              #else
+                pair[0];
+              #endif
+            *ptr0 = pair1;
+
+            _distances[-1] = (UInt32)(d - _distances);
+
+            #ifdef USE_LONG_MATCH_OPT
+
+                if (hash == size || *hash != delta || lenLimit[diff] != lenLimit[0] || d >= limit)
+                  break;
+
+            {
+              for (;;)
+              {
+                hash++;
+                pos++;
+                cur++;
+                lenLimit++;
+                {
+                  CLzRef *ptr = son + ((size_t)(pos) << 1) - CYC_TO_POS_OFFSET * 2;
+                  #if 0
+                  *(UInt64 *)(void *)ptr = ((const UInt64 *)(const void *)ptr)[diff];
+                  #else
+                  const UInt32 p0 = ptr[0 + (diff * 2)];
+                  const UInt32 p1 = ptr[1 + (diff * 2)];
+                  ptr[0] = p0;
+                  ptr[1] = p1;
+                  // ptr[0] = ptr[0 + (diff * 2)];
+                  // ptr[1] = ptr[1 + (diff * 2)];
+                  #endif
+                }
+                // PrintSon(son + 2, pos - 1);
+                // printf("\npos = %x delta = %x\n", pos, delta);
+                len++;
+                *d++ = 2;
+                *d++ = (UInt32)(len - cur);
+                *d++ = delta - 1;
+                if (hash == size || *hash != delta || lenLimit[diff] != lenLimit[0] || d >= limit)
+                  break;
+              }
+            }
+            #endif
+
+            break;
+          }
+        }
+      }
+
+      {
+        const UInt32 curMatch = (UInt32)pos - delta; // (UInt32)(pos + diff);
+        if (len[diff] < len[0])
+        {
+          delta = pair[1];
+          if (delta >= curMatch)
+            return NULL;
+          *ptr1 = curMatch;
+          ptr1 = pair + 1;
+          len1 = len;
+        }
+        else
+        {
+          delta = *pair;
+          if (delta >= curMatch)
+            return NULL;
+          *ptr0 = curMatch;
+          ptr0 = pair;
+          len0 = len;
+        }
+
+        delta = (UInt32)pos - delta;
+ 
+        if (--cutValue == 0 || delta >= pos)
+        {
+          *ptr0 = *ptr1 = kEmptyHashValue;
+          _distances[-1] = (UInt32)(d - _distances);
+          break;
+        }
+      }
+    }
+  } // for (tree iterations)
+}
+    pos++;
+    cur++;
+  }
+  while (d < limit);
+  *posRes = (UInt32)pos;
+  return d;
+}
+*/
+
+/* define cbs if you use 2 functions.
+       GetMatchesSpecN_1() :  (pos <  _cyclicBufferSize)
+       GetMatchesSpecN_2() :  (pos >= _cyclicBufferSize)
+
+  do not define cbs if you use 1 function:
+       GetMatchesSpecN_2()
+*/
+
+// #define cbs _cyclicBufferSize
+
+/*
+  we use size_t for (pos) and (_cyclicBufferPos_ instead of UInt32
+  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 _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,
+    UInt32 _cutValue, UInt32 *d, size_t _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size,
+    size_t _cyclicBufferPos, UInt32 _cyclicBufferSize,
+    UInt32 *posRes)
+{
+  do // while (hash != size)
+  {
+    UInt32 delta;
+    
+  #ifndef cbs
+    UInt32 cbs;
+  #endif
+
+    if (hash == size)
+      break;
+
+    delta = *hash++;
+
+    if (delta == 0)
+      return NULL;
+
+    lenLimit++;
+
+  #ifndef cbs
+    cbs = _cyclicBufferSize;
+    if ((UInt32)pos < cbs)
+    {
+      if (delta > (UInt32)pos)
+        return NULL;
+      cbs = (UInt32)pos;
+    }
+  #endif
+
+    if (delta >= cbs)
+    {
+      CLzRef *ptr1 = son + ((size_t)_cyclicBufferPos << 1);
+      *d++ = 0;
+      ptr1[0] = kEmptyHashValue;
+      ptr1[1] = kEmptyHashValue;
+    }
+else
+{
+  UInt32 *_distances = ++d;
+
+  CLzRef *ptr0 = son + ((size_t)_cyclicBufferPos << 1) + 1;
+  CLzRef *ptr1 = son + ((size_t)_cyclicBufferPos << 1);
+
+  UInt32 cutValue = _cutValue;
+  const Byte *len0 = cur, *len1 = cur;
+  const Byte *maxLen = cur + _maxLen;
+
+  // if (cutValue == 0) { *ptr0 = *ptr1 = kEmptyHashValue; } else
+  for (LOG_ITER(g_NumIters_Tree++);;)
+  {
+    LOG_ITER(g_NumIters_Loop++);
+    {
+      // SPEC code
+      CLzRef *pair = son + ((size_t)((ptrdiff_t)_cyclicBufferPos - (ptrdiff_t)delta
+          + (ptrdiff_t)(UInt32)(_cyclicBufferPos < delta ? cbs : 0)
+          ) << 1);
+
+      const ptrdiff_t diff = (ptrdiff_t)0 - (ptrdiff_t)delta;
+      const Byte *len = (len0 < len1 ? len0 : len1);
+
+    #ifdef USE_SON_PREFETCH
+      const UInt32 pair0 = *pair;
+    #endif
+
+      if (len[diff] == len[0])
+      {
+        if (++len != lenLimit && len[diff] == len[0])
+          while (++len != lenLimit)
+          {
+            LOG_ITER(g_NumIters_Bytes++);
+            if (len[diff] != len[0])
+              break;
+          }
+        if (maxLen < len)
+        {
+          maxLen = len;
+          *d++ = (UInt32)(len - cur);
+          *d++ = delta - 1;
+          
+          if (len == lenLimit)
+          {
+            const UInt32 pair1 = pair[1];
+            *ptr1 =
+              #ifdef USE_SON_PREFETCH
+                pair0;
+              #else
+                pair[0];
+              #endif
+            *ptr0 = pair1;
+
+            _distances[-1] = (UInt32)(d - _distances);
+
+            #ifdef USE_LONG_MATCH_OPT
+
+                if (hash == size || *hash != delta || lenLimit[diff] != lenLimit[0] || d >= limit)
+                  break;
+
+            {
+              for (;;)
+              {
+                *d++ = 2;
+                *d++ = (UInt32)(lenLimit - cur);
+                *d++ = delta - 1;
+                cur++;
+                lenLimit++;
+                // SPEC
+                _cyclicBufferPos++;
+                {
+                  // SPEC code
+                  CLzRef *dest = son + ((size_t)(_cyclicBufferPos) << 1);
+                  const CLzRef *src = dest + ((diff
+                      + (ptrdiff_t)(UInt32)((_cyclicBufferPos < delta) ? cbs : 0)) << 1);
+                  // CLzRef *ptr = son + ((size_t)(pos) << 1) - CYC_TO_POS_OFFSET * 2;
+                  #if 0
+                  *(UInt64 *)(void *)dest = *((const UInt64 *)(const void *)src);
+                  #else
+                  const UInt32 p0 = src[0];
+                  const UInt32 p1 = src[1];
+                  dest[0] = p0;
+                  dest[1] = p1;
+                  #endif
+                }
+                pos++;
+                hash++;
+                if (hash == size || *hash != delta || lenLimit[diff] != lenLimit[0] || d >= limit)
+                  break;
+              } // for() end for long matches
+            }
+            #endif
+
+            break; // break from TREE iterations
+          }
+        }
+      }
+      {
+        const UInt32 curMatch = (UInt32)pos - delta; // (UInt32)(pos + diff);
+        if (len[diff] < len[0])
+        {
+          delta = pair[1];
+          *ptr1 = curMatch;
+          ptr1 = pair + 1;
+          len1 = len;
+          if (delta >= curMatch)
+            return NULL;
+        }
+        else
+        {
+          delta = *pair;
+          *ptr0 = curMatch;
+          ptr0 = pair;
+          len0 = len;
+          if (delta >= curMatch)
+            return NULL;
+        }
+        delta = (UInt32)pos - delta;
+ 
+        if (--cutValue == 0 || delta >= cbs)
+        {
+          *ptr0 = *ptr1 = kEmptyHashValue;
+          _distances[-1] = (UInt32)(d - _distances);
+          break;
+        }
+      }
+    }
+  } // for (tree iterations)
+}
+    pos++;
+    _cyclicBufferPos++;
+    cur++;
+  }
+  while (d < limit);
+  *posRes = (UInt32)pos;
+  return d;
+}
+
+
+
+/*
+typedef UInt32 uint32plus; // size_t
+
+UInt32 * MY_FAST_CALL 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)
+{
+  do // while (hash != size)
+  {
+    UInt32 delta;
+
+  #ifndef cbs
+    UInt32 cbs;
+  #endif
+
+    if (hash == size)
+      break;
+
+    delta = *hash++;
+
+    if (delta == 0)
+      return NULL;
+
+  #ifndef cbs
+    cbs = _cyclicBufferSize;
+    if ((UInt32)pos < cbs)
+    {
+      if (delta > (UInt32)pos)
+        return NULL;
+      cbs = (UInt32)pos;
+    }
+  #endif
+    
+    if (delta >= cbs)
+    {
+      CLzRef *ptr1 = son + ((size_t)_cyclicBufferPos << 1);
+      *d++ = 0;
+      ptr1[0] = kEmptyHashValue;
+      ptr1[1] = kEmptyHashValue;
+    }
+else
+{
+  CLzRef *ptr0 = son + ((size_t)_cyclicBufferPos << 1) + 1;
+  CLzRef *ptr1 = son + ((size_t)_cyclicBufferPos << 1);
+  UInt32 *_distances = ++d;
+  uint32plus len0 = 0, len1 = 0;
+  UInt32 cutValue = _cutValue;
+  uint32plus maxLen = _maxLen;
+  // lenLimit++; // const Byte *lenLimit = cur + _lenLimit;
+
+  for (LOG_ITER(g_NumIters_Tree++);;)
+  {
+    LOG_ITER(g_NumIters_Loop++);
+    {
+      // const ptrdiff_t diff = (ptrdiff_t)0 - (ptrdiff_t)delta;
+      CLzRef *pair = son + ((size_t)((ptrdiff_t)_cyclicBufferPos - delta
+          + (ptrdiff_t)(UInt32)(_cyclicBufferPos < delta ? cbs : 0)
+          ) << 1);
+      const Byte *pb = cur - delta;
+      uint32plus len = (len0 < len1 ? len0 : len1);
+
+    #ifdef USE_SON_PREFETCH
+      const UInt32 pair0 = *pair;
+    #endif
+
+      if (pb[len] == cur[len])
+      {
+        if (++len != lenLimit && pb[len] == cur[len])
+          while (++len != lenLimit)
+            if (pb[len] != cur[len])
+              break;
+        if (maxLen < len)
+        {
+          maxLen = len;
+          *d++ = (UInt32)len;
+          *d++ = delta - 1;
+          if (len == lenLimit)
+          {
+            {
+              const UInt32 pair1 = pair[1];
+              *ptr0 = pair1;
+              *ptr1 =
+              #ifdef USE_SON_PREFETCH
+                pair0;
+              #else
+                pair[0];
+              #endif
+            }
+
+            _distances[-1] = (UInt32)(d - _distances);
+
+            #ifdef USE_LONG_MATCH_OPT
+
+                if (hash == size || *hash != delta || pb[lenLimit] != cur[lenLimit] || d >= limit)
+                  break;
+
+            {
+              const ptrdiff_t diff = (ptrdiff_t)0 - (ptrdiff_t)delta;
+              for (;;)
+              {
+                *d++ = 2;
+                *d++ = (UInt32)lenLimit;
+                *d++ = delta - 1;
+                _cyclicBufferPos++;
+                {
+                  CLzRef *dest = son + ((size_t)_cyclicBufferPos << 1);
+                  const CLzRef *src = dest + ((diff +
+                      (ptrdiff_t)(UInt32)(_cyclicBufferPos < delta ? cbs : 0)) << 1);
+                #if 0
+                  *(UInt64 *)(void *)dest = *((const UInt64 *)(const void *)src);
+                #else
+                  const UInt32 p0 = src[0];
+                  const UInt32 p1 = src[1];
+                  dest[0] = p0;
+                  dest[1] = p1;
+                #endif
+                }
+                hash++;
+                pos++;
+                cur++;
+                pb++;
+                if (hash == size || *hash != delta || pb[lenLimit] != cur[lenLimit] || d >= limit)
+                  break;
+              }
+            }
+            #endif
+
+            break;
+          }
+        }
+      }
+      {
+        const UInt32 curMatch = (UInt32)pos - delta;
+        if (pb[len] < cur[len])
+        {
+          delta = pair[1];
+          *ptr1 = curMatch;
+          ptr1 = pair + 1;
+          len1 = len;
+        }
+        else
+        {
+          delta = *pair;
+          *ptr0 = curMatch;
+          ptr0 = pair;
+          len0 = len;
+        }
+
+        {
+          if (delta >= curMatch)
+            return NULL;
+          delta = (UInt32)pos - delta;
+          if (delta >= cbs
+              // delta >= _cyclicBufferSize || delta >= pos
+              || --cutValue == 0)
+          {
+            *ptr0 = *ptr1 = kEmptyHashValue;
+            _distances[-1] = (UInt32)(d - _distances);
+            break;
+          }
+        }
+      }
+    }
+  } // for (tree iterations)
+}
+    pos++;
+    _cyclicBufferPos++;
+    cur++;
+  }
+  while (d < limit);
+  *posRes = (UInt32)pos;
+  return d;
+}
+*/

libraries/lzma/C/LzHash.h

@@ -1,57 +1,34 @@
 /* LzHash.h -- HASH functions for LZ algorithms
-2015-04-12 : Igor Pavlov : Public domain */
+2019-10-30 : Igor Pavlov : Public domain */
 
 #ifndef __LZ_HASH_H
 #define __LZ_HASH_H
 
+/*
+  (kHash2Size >= (1 <<  8)) : Required
+  (kHash3Size >= (1 << 16)) : Required
+*/
+
 #define kHash2Size (1 << 10)
 #define kHash3Size (1 << 16)
-#define kHash4Size (1 << 20)
+// #define kHash4Size (1 << 20)
 
 #define kFix3HashSize (kHash2Size)
 #define kFix4HashSize (kHash2Size + kHash3Size)
-#define kFix5HashSize (kHash2Size + kHash3Size + kHash4Size)
+// #define kFix5HashSize (kHash2Size + kHash3Size + kHash4Size)
 
-#define HASH2_CALC hv = cur[0] | ((UInt32)cur[1] << 8);
+/*
+  We use up to 3 crc values for hash:
+    crc0
+    crc1 << Shift_1
+    crc2 << Shift_2
+  (Shift_1 = 5) and (Shift_2 = 10) is good tradeoff.
+  Small values for Shift are not good for collision rate.
+  Big value for Shift_2 increases the minimum size
+  of hash table, that will be slow for small files.
+*/
 
-#define HASH3_CALC { \
+#define kLzHash_CrcShift_1 5
-  UInt32 temp = p->crc[cur[0]] ^ cur[1]; \
+#define kLzHash_CrcShift_2 10
-  h2 = temp & (kHash2Size - 1); \
-  hv = (temp ^ ((UInt32)cur[2] << 8)) & p->hashMask; }
-
-#define HASH4_CALC { \
-  UInt32 temp = p->crc[cur[0]] ^ cur[1]; \
-  h2 = temp & (kHash2Size - 1); \
-  temp ^= ((UInt32)cur[2] << 8); \
-  h3 = temp & (kHash3Size - 1); \
-  hv = (temp ^ (p->crc[cur[3]] << 5)) & p->hashMask; }
-
-#define HASH5_CALC { \
-  UInt32 temp = p->crc[cur[0]] ^ cur[1]; \
-  h2 = temp & (kHash2Size - 1); \
-  temp ^= ((UInt32)cur[2] << 8); \
-  h3 = temp & (kHash3Size - 1); \
-  temp ^= (p->crc[cur[3]] << 5); \
-  h4 = temp & (kHash4Size - 1); \
-  hv = (temp ^ (p->crc[cur[4]] << 3)) & p->hashMask; }
-
-/* #define HASH_ZIP_CALC hv = ((cur[0] | ((UInt32)cur[1] << 8)) ^ p->crc[cur[2]]) & 0xFFFF; */
-#define HASH_ZIP_CALC hv = ((cur[2] | ((UInt32)cur[0] << 8)) ^ p->crc[cur[1]]) & 0xFFFF;
-
-
-#define MT_HASH2_CALC \
-  h2 = (p->crc[cur[0]] ^ cur[1]) & (kHash2Size - 1);
-
-#define MT_HASH3_CALC { \
-  UInt32 temp = p->crc[cur[0]] ^ cur[1]; \
-  h2 = temp & (kHash2Size - 1); \
-  h3 = (temp ^ ((UInt32)cur[2] << 8)) & (kHash3Size - 1); }
-
-#define MT_HASH4_CALC { \
-  UInt32 temp = p->crc[cur[0]] ^ cur[1]; \
-  h2 = temp & (kHash2Size - 1); \
-  temp ^= ((UInt32)cur[2] << 8); \
-  h3 = temp & (kHash3Size - 1); \
-  h4 = (temp ^ (p->crc[cur[3]] << 5)) & (kHash4Size - 1); }
 
 #endif

libraries/lzma/C/Lzma2Dec.c

@@ -1,5 +1,5 @@
 /* Lzma2Dec.c -- LZMA2 Decoder
-2019-02-02 : Igor Pavlov : Public domain */
+2021-02-09 : Igor Pavlov : Public domain */
 
 /* #define SHOW_DEBUG_INFO */
 
@@ -93,7 +93,8 @@ void Lzma2Dec_Init(CLzma2Dec *p)
   LzmaDec_Init(&p->decoder);
 }
 
-static ELzma2State Lzma2Dec_UpdateState(CLzma2Dec *p, Byte b)
+// ELzma2State
+static unsigned Lzma2Dec_UpdateState(CLzma2Dec *p, Byte b)
 {
   switch (p->state)
   {

libraries/lzma/C/LzmaDec.c

@@ -1,5 +1,5 @@
 /* LzmaDec.c -- LZMA Decoder
-2018-07-04 : Igor Pavlov : Public domain */
+2021-04-01 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -13,10 +13,12 @@
 
 #define kNumBitModelTotalBits 11
 #define kBitModelTotal (1 << kNumBitModelTotalBits)
-#define kNumMoveBits 5
 
 #define RC_INIT_SIZE 5
 
+#ifndef _LZMA_DEC_OPT
+
+#define kNumMoveBits 5
 #define NORMALIZE if (range < kTopValue) { range <<= 8; code = (code << 8) | (*buf++); }
 
 #define IF_BIT_0(p) ttt = *(p); NORMALIZE; bound = (range >> kNumBitModelTotalBits) * (UInt32)ttt; if (code < bound)
@@ -62,9 +64,10 @@
   probLit = prob + (offs + bit + symbol); \
   GET_BIT2(probLit, symbol, offs ^= bit; , ;)
 
+#endif // _LZMA_DEC_OPT
 
 
-#define NORMALIZE_CHECK if (range < kTopValue) { if (buf >= bufLimit) return DUMMY_ERROR; range <<= 8; code = (code << 8) | (*buf++); }
+#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 UPDATE_0_CHECK range = bound;
@@ -114,6 +117,9 @@
 #define kMatchMinLen 2
 #define kMatchSpecLenStart (kMatchMinLen + kLenNumLowSymbols * 2 + kLenNumHighSymbols)
 
+#define kMatchSpecLen_Error_Data (1 << 9)
+#define kMatchSpecLen_Error_Fail (kMatchSpecLen_Error_Data - 1)
+
 /* External ASM code needs same CLzmaProb array layout. So don't change it. */
 
 /* (probs_1664) is faster and better for code size at some platforms */
@@ -166,10 +172,12 @@
 
 /*
 p->remainLen : shows status of LZMA decoder:
-    < kMatchSpecLenStart : normal remain
+    < kMatchSpecLenStart  : the number of bytes to be copied with (p->rep0) offset
-    = kMatchSpecLenStart : finished
+    = kMatchSpecLenStart  : the LZMA stream was finished with end mark
-    = kMatchSpecLenStart + 1 : need init range coder
+    = kMatchSpecLenStart + 1  : need init range coder
-    = kMatchSpecLenStart + 2 : need init range coder and state
+    = kMatchSpecLenStart + 2  : need init range coder and state
+    = kMatchSpecLen_Error_Fail                : Internal Code Failure
+    = kMatchSpecLen_Error_Data + [0 ... 273]  : LZMA Data Error
 */
 
 /* ---------- LZMA_DECODE_REAL ---------- */
@@ -188,23 +196,31 @@ In:
   {
     LzmaDec_TryDummy() was called before to exclude LITERAL and MATCH-REP cases.
     So first symbol can be only MATCH-NON-REP. And if that MATCH-NON-REP symbol
-    is not END_OF_PAYALOAD_MARKER, then function returns error code.
+    is not END_OF_PAYALOAD_MARKER, then the function doesn't write any byte to dictionary,
+    the function returns SZ_OK, and the caller can use (p->remainLen) and (p->reps[0]) later.
   }
 
 Processing:
-  first LZMA symbol will be decoded in any case
+  The first LZMA symbol will be decoded in any case.
-  All checks for limits are at the end of main loop,
+  All main checks for limits are at the end of main loop,
-  It will decode new LZMA-symbols while (p->buf < bufLimit && dicPos < limit),
+  It decodes additional LZMA-symbols while (p->buf < bufLimit && dicPos < limit),
   RangeCoder is still without last normalization when (p->buf < bufLimit) is being checked.
+  But if (p->buf < bufLimit), the caller provided at least (LZMA_REQUIRED_INPUT_MAX + 1) bytes for
+  next iteration  before limit (bufLimit + LZMA_REQUIRED_INPUT_MAX),
+  that is enough for worst case LZMA symbol with one additional RangeCoder normalization for one bit.
+  So that function never reads bufLimit [LZMA_REQUIRED_INPUT_MAX] byte.
 
 Out:
   RangeCoder is normalized
   Result:
     SZ_OK - OK
-    SZ_ERROR_DATA - Error
+      p->remainLen:
-  p->remainLen:
+        < kMatchSpecLenStart : the number of bytes to be copied with (p->reps[0]) offset
-    < kMatchSpecLenStart : normal remain
+        = kMatchSpecLenStart : the LZMA stream was finished with end mark
-    = kMatchSpecLenStart : finished
+
+    SZ_ERROR_DATA - error, when the MATCH-Symbol refers out of dictionary
+      p->remainLen : undefined
+      p->reps[*]    : undefined
 */
 
 
@@ -316,11 +332,6 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit
       else
       {
         UPDATE_1(prob);
-        /*
-        // that case was checked before with kBadRepCode
-        if (checkDicSize == 0 && processedPos == 0)
-          return SZ_ERROR_DATA;
-        */
         prob = probs + IsRepG0 + state;
         IF_BIT_0(prob)
         {
@@ -329,6 +340,13 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit
           IF_BIT_0(prob)
           {
             UPDATE_0(prob);
+  
+            // that case was checked before with kBadRepCode
+            // if (checkDicSize == 0 && processedPos == 0) { len = kMatchSpecLen_Error_Data + 1; break; }
+            // The caller doesn't allow (dicPos == limit) case here
+            // so we don't need the following check:
+            // if (dicPos == limit) { state = state < kNumLitStates ? 9 : 11; len = 1; break; }
+            
             dic[dicPos] = dic[dicPos - rep0 + (dicPos < rep0 ? dicBufSize : 0)];
             dicPos++;
             processedPos++;
@@ -518,8 +536,10 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit
         state = (state < kNumStates + kNumLitStates) ? kNumLitStates : kNumLitStates + 3;
         if (distance >= (checkDicSize == 0 ? processedPos: checkDicSize))
         {
-          p->dicPos = dicPos;
+          len += kMatchSpecLen_Error_Data + kMatchMinLen;
-          return SZ_ERROR_DATA;
+          // len = kMatchSpecLen_Error_Data;
+          // len += kMatchMinLen;
+          break;
         }
       }
 
@@ -532,8 +552,13 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit
         
         if ((rem = limit - dicPos) == 0)
         {
-          p->dicPos = dicPos;
+          /*
-          return SZ_ERROR_DATA;
+          We stop decoding and return SZ_OK, and we can resume decoding later.
+          Any error conditions can be tested later in caller code.
+          For more strict mode we can stop decoding with error
+          // len += kMatchSpecLen_Error_Data;
+          */
+          break;
         }
         
         curLen = ((rem < len) ? (unsigned)rem : len);
@@ -572,7 +597,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit
   p->buf = buf;
   p->range = range;
   p->code = code;
-  p->remainLen = (UInt32)len;
+  p->remainLen = (UInt32)len; // & (kMatchSpecLen_Error_Data - 1); // we can write real length for error matches too.
   p->dicPos = dicPos;
   p->processedPos = processedPos;
   p->reps[0] = rep0;
@@ -580,40 +605,61 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit
   p->reps[2] = rep2;
   p->reps[3] = rep3;
   p->state = (UInt32)state;
-
+  if (len >= kMatchSpecLen_Error_Data)
+    return SZ_ERROR_DATA;
   return SZ_OK;
 }
 #endif
 
+
+
 static void MY_FAST_CALL LzmaDec_WriteRem(CLzmaDec *p, SizeT limit)
 {
-  if (p->remainLen != 0 && p->remainLen < kMatchSpecLenStart)
+  unsigned len = (unsigned)p->remainLen;
+  if (len == 0 /* || len >= kMatchSpecLenStart */)
+    return;
   {
-    Byte *dic = p->dic;
     SizeT dicPos = p->dicPos;
-    SizeT dicBufSize = p->dicBufSize;
+    Byte *dic;
-    unsigned len = (unsigned)p->remainLen;
+    SizeT dicBufSize;
-    SizeT rep0 = p->reps[0]; /* we use SizeT to avoid the BUG of VC14 for AMD64 */
+    SizeT rep0;   /* we use SizeT to avoid the BUG of VC14 for AMD64 */
-    SizeT rem = limit - dicPos;
+    {
-    if (rem < len)
+      SizeT rem = limit - dicPos;
-      len = (unsigned)(rem);
+      if (rem < len)
+      {
+        len = (unsigned)(rem);
+        if (len == 0)
+          return;
+      }
+    }
 
     if (p->checkDicSize == 0 && p->prop.dicSize - p->processedPos <= len)
       p->checkDicSize = p->prop.dicSize;
 
     p->processedPos += (UInt32)len;
     p->remainLen -= (UInt32)len;
-    while (len != 0)
+    dic = p->dic;
+    rep0 = p->reps[0];
+    dicBufSize = p->dicBufSize;
+    do
     {
-      len--;
       dic[dicPos] = dic[dicPos - rep0 + (dicPos < rep0 ? dicBufSize : 0)];
       dicPos++;
     }
+    while (--len);
     p->dicPos = dicPos;
   }
 }
 
 
+/*
+At staring of new stream we have one of the following symbols:
+  - Literal        - is allowed
+  - Non-Rep-Match  - is allowed only if it's end marker symbol
+  - Rep-Match      - is not allowed
+We use early check of (RangeCoder:Code) over kBadRepCode to simplify main decoding code
+*/
+
 #define kRange0 0xFFFFFFFF
 #define kBound0 ((kRange0 >> kNumBitModelTotalBits) << (kNumBitModelTotalBits - 1))
 #define kBadRepCode (kBound0 + (((kRange0 - kBound0) >> kNumBitModelTotalBits) << (kNumBitModelTotalBits - 1)))
@@ -621,69 +667,77 @@ static void MY_FAST_CALL LzmaDec_WriteRem(CLzmaDec *p, SizeT limit)
   #error Stop_Compiling_Bad_LZMA_Check
 #endif
 
+
+/*
+LzmaDec_DecodeReal2():
+  It calls LZMA_DECODE_REAL() and it adjusts limit according (p->checkDicSize).
+
+We correct (p->checkDicSize) after LZMA_DECODE_REAL() and in LzmaDec_WriteRem(),
+and we support the following state of (p->checkDicSize):
+  if (total_processed < p->prop.dicSize) then
+  {
+    (total_processed == p->processedPos)
+    (p->checkDicSize == 0)
+  }
+  else
+    (p->checkDicSize == p->prop.dicSize)
+*/
+
 static int MY_FAST_CALL LzmaDec_DecodeReal2(CLzmaDec *p, SizeT limit, const Byte *bufLimit)
 {
-  do
+  if (p->checkDicSize == 0)
   {
-    SizeT limit2 = limit;
+    UInt32 rem = p->prop.dicSize - p->processedPos;
-    if (p->checkDicSize == 0)
+    if (limit - p->dicPos > rem)
-    {
+      limit = p->dicPos + rem;
-      UInt32 rem = p->prop.dicSize - p->processedPos;
+  }
-      if (limit - p->dicPos > rem)
+  {
-        limit2 = p->dicPos + rem;
+    int res = LZMA_DECODE_REAL(p, limit, bufLimit);
-
-      if (p->processedPos == 0)
-        if (p->code >= kBadRepCode)
-          return SZ_ERROR_DATA;
-    }
-
-    RINOK(LZMA_DECODE_REAL(p, limit2, bufLimit));
-    
     if (p->checkDicSize == 0 && p->processedPos >= p->prop.dicSize)
       p->checkDicSize = p->prop.dicSize;
-    
+    return res;
-    LzmaDec_WriteRem(p, limit);
   }
-  while (p->dicPos < limit && p->buf < bufLimit && p->remainLen < kMatchSpecLenStart);
-
-  return 0;
 }
 
+
+
 typedef enum
 {
-  DUMMY_ERROR, /* unexpected end of input stream */
+  DUMMY_INPUT_EOF, /* need more input data */
   DUMMY_LIT,
   DUMMY_MATCH,
   DUMMY_REP
 } ELzmaDummy;
 
-static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, SizeT inSize)
+
+#define IS_DUMMY_END_MARKER_POSSIBLE(dummyRes) ((dummyRes) == DUMMY_MATCH)
+
+static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, const Byte **bufOut)
 {
   UInt32 range = p->range;
   UInt32 code = p->code;
-  const Byte *bufLimit = buf + inSize;
+  const Byte *bufLimit = *bufOut;
   const CLzmaProb *probs = GET_PROBS;
   unsigned state = (unsigned)p->state;
   ELzmaDummy res;
 
+  for (;;)
   {
     const CLzmaProb *prob;
     UInt32 bound;
     unsigned ttt;
-    unsigned posState = CALC_POS_STATE(p->processedPos, (1 << p->prop.pb) - 1);
+    unsigned posState = CALC_POS_STATE(p->processedPos, ((unsigned)1 << p->prop.pb) - 1);
 
     prob = probs + IsMatch + COMBINED_PS_STATE;
     IF_BIT_0_CHECK(prob)
     {
       UPDATE_0_CHECK
 
-      /* if (bufLimit - buf >= 7) return DUMMY_LIT; */
-
       prob = probs + Literal;
       if (p->checkDicSize != 0 || p->processedPos != 0)
         prob += ((UInt32)LZMA_LIT_SIZE *
-            ((((p->processedPos) & ((1 << (p->prop.lp)) - 1)) << p->prop.lc) +
+            ((((p->processedPos) & (((unsigned)1 << (p->prop.lp)) - 1)) << p->prop.lc) +
-            (p->dic[(p->dicPos == 0 ? p->dicBufSize : p->dicPos) - 1] >> (8 - p->prop.lc))));
+            ((unsigned)p->dic[(p->dicPos == 0 ? p->dicBufSize : p->dicPos) - 1] >> (8 - p->prop.lc))));
 
       if (state < kNumLitStates)
       {
@@ -735,8 +789,7 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, SizeT inS
           IF_BIT_0_CHECK(prob)
           {
             UPDATE_0_CHECK;
-            NORMALIZE_CHECK;
+            break;
-            return DUMMY_REP;
           }
           else
           {
@@ -812,8 +865,6 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, SizeT inS
         {
           unsigned numDirectBits = ((posSlot >> 1) - 1);
 
-          /* if (bufLimit - buf >= 8) return DUMMY_MATCH; */
-
           if (posSlot < kEndPosModelIndex)
           {
             prob = probs + SpecPos + ((2 | (posSlot & 1)) << numDirectBits);
@@ -844,12 +895,15 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, SizeT inS
         }
       }
     }
+    break;
   }
   NORMALIZE_CHECK;
+
+  *bufOut = buf;
   return res;
 }
 
-
+void LzmaDec_InitDicAndState(CLzmaDec *p, BoolInt initDic, BoolInt initState);
 void LzmaDec_InitDicAndState(CLzmaDec *p, BoolInt initDic, BoolInt initState)
 {
   p->remainLen = kMatchSpecLenStart + 1;
@@ -872,16 +926,41 @@ void LzmaDec_Init(CLzmaDec *p)
 }
 
 
+/*
+LZMA supports optional end_marker.
+So the decoder can lookahead for one additional LZMA-Symbol to check end_marker.
+That additional LZMA-Symbol can require up to LZMA_REQUIRED_INPUT_MAX bytes in input stream.
+When the decoder reaches dicLimit, it looks (finishMode) parameter:
+  if (finishMode == LZMA_FINISH_ANY), the decoder doesn't lookahead
+  if (finishMode != LZMA_FINISH_ANY), the decoder lookahead, if end_marker is possible for current position
+
+When the decoder lookahead, and the lookahead symbol is not end_marker, we have two ways:
+  1) Strict mode (default) : the decoder returns SZ_ERROR_DATA.
+  2) The relaxed mode (alternative mode) : we could return SZ_OK, and the caller
+     must check (status) value. The caller can show the error,
+     if the end of stream is expected, and the (status) is noit
+     LZMA_STATUS_FINISHED_WITH_MARK or LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK.
+*/
+
+
+#define RETURN__NOT_FINISHED__FOR_FINISH \
+  *status = LZMA_STATUS_NOT_FINISHED; \
+  return SZ_ERROR_DATA; // for strict mode
+  // return SZ_OK; // for relaxed mode
+
+
 SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *srcLen,
     ELzmaFinishMode finishMode, ELzmaStatus *status)
 {
   SizeT inSize = *srcLen;
   (*srcLen) = 0;
-  
   *status = LZMA_STATUS_NOT_SPECIFIED;
 
   if (p->remainLen > kMatchSpecLenStart)
   {
+    if (p->remainLen > kMatchSpecLenStart + 2)
+      return p->remainLen == kMatchSpecLen_Error_Fail ? SZ_ERROR_FAIL : SZ_ERROR_DATA;
+
     for (; inSize > 0 && p->tempBufSize < RC_INIT_SIZE; (*srcLen)++, inSize--)
       p->tempBuf[p->tempBufSize++] = *src++;
     if (p->tempBufSize != 0 && p->tempBuf[0] != 0)
@@ -896,6 +975,12 @@ SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *sr
       | ((UInt32)p->tempBuf[2] << 16)
       | ((UInt32)p->tempBuf[3] << 8)
       | ((UInt32)p->tempBuf[4]);
+
+    if (p->checkDicSize == 0
+        && p->processedPos == 0
+        && p->code >= kBadRepCode)
+      return SZ_ERROR_DATA;
+
     p->range = 0xFFFFFFFF;
     p->tempBufSize = 0;
 
@@ -913,10 +998,21 @@ SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *sr
     p->remainLen = 0;
   }
 
-  LzmaDec_WriteRem(p, dicLimit);
+  for (;;)
-
-  while (p->remainLen != kMatchSpecLenStart)
   {
+    if (p->remainLen == kMatchSpecLenStart)
+    {
+      if (p->code != 0)
+        return SZ_ERROR_DATA;
+      *status = LZMA_STATUS_FINISHED_WITH_MARK;
+      return SZ_OK;
+    }
+
+    LzmaDec_WriteRem(p, dicLimit);
+
+    {
+      // (p->remainLen == 0 || p->dicPos == dicLimit)
+
       int checkEndMarkNow = 0;
 
       if (p->dicPos >= dicLimit)
@@ -933,92 +1029,174 @@ SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *sr
         }
         if (p->remainLen != 0)
         {
-          *status = LZMA_STATUS_NOT_FINISHED;
+          RETURN__NOT_FINISHED__FOR_FINISH;
-          return SZ_ERROR_DATA;
         }
         checkEndMarkNow = 1;
       }
 
+      // (p->remainLen == 0)
+
       if (p->tempBufSize == 0)
       {
-        SizeT processed;
         const Byte *bufLimit;
+        int dummyProcessed = -1;
+        
         if (inSize < LZMA_REQUIRED_INPUT_MAX || checkEndMarkNow)
         {
-          int dummyRes = LzmaDec_TryDummy(p, src, inSize);
+          const Byte *bufOut = src + inSize;
-          if (dummyRes == DUMMY_ERROR)
+          
+          ELzmaDummy dummyRes = LzmaDec_TryDummy(p, src, &bufOut);
+          
+          if (dummyRes == DUMMY_INPUT_EOF)
           {
-            memcpy(p->tempBuf, src, inSize);
+            size_t i;
-            p->tempBufSize = (unsigned)inSize;
+            if (inSize >= LZMA_REQUIRED_INPUT_MAX)
+              break;
             (*srcLen) += inSize;
+            p->tempBufSize = (unsigned)inSize;
+            for (i = 0; i < inSize; i++)
+              p->tempBuf[i] = src[i];
             *status = LZMA_STATUS_NEEDS_MORE_INPUT;
             return SZ_OK;
           }
-          if (checkEndMarkNow && dummyRes != DUMMY_MATCH)
+ 
+          dummyProcessed = (int)(bufOut - src);
+          if ((unsigned)dummyProcessed > LZMA_REQUIRED_INPUT_MAX)
+            break;
+          
+          if (checkEndMarkNow && !IS_DUMMY_END_MARKER_POSSIBLE(dummyRes))
           {
-            *status = LZMA_STATUS_NOT_FINISHED;
+            unsigned i;
-            return SZ_ERROR_DATA;
+            (*srcLen) += (unsigned)dummyProcessed;
+            p->tempBufSize = (unsigned)dummyProcessed;
+            for (i = 0; i < (unsigned)dummyProcessed; i++)
+              p->tempBuf[i] = src[i];
+            // p->remainLen = kMatchSpecLen_Error_Data;
+            RETURN__NOT_FINISHED__FOR_FINISH;
           }
+          
           bufLimit = src;
+          // we will decode only one iteration
         }
         else
           bufLimit = src + inSize - LZMA_REQUIRED_INPUT_MAX;
+
         p->buf = src;
-        if (LzmaDec_DecodeReal2(p, dicLimit, bufLimit) != 0)
+        
-          return SZ_ERROR_DATA;
-        processed = (SizeT)(p->buf - src);
-        (*srcLen) += processed;
-        src += processed;
-        inSize -= processed;
-      }
-      else
-      {
-        unsigned rem = p->tempBufSize, lookAhead = 0;
-        while (rem < LZMA_REQUIRED_INPUT_MAX && lookAhead < inSize)
-          p->tempBuf[rem++] = src[lookAhead++];
-        p->tempBufSize = rem;
-        if (rem < LZMA_REQUIRED_INPUT_MAX || checkEndMarkNow)
         {
-          int dummyRes = LzmaDec_TryDummy(p, p->tempBuf, (SizeT)rem);
+          int res = LzmaDec_DecodeReal2(p, dicLimit, bufLimit);
-          if (dummyRes == DUMMY_ERROR)
+          
+          SizeT processed = (SizeT)(p->buf - src);
+
+          if (dummyProcessed < 0)
           {
-            (*srcLen) += (SizeT)lookAhead;
+            if (processed > inSize)
-            *status = LZMA_STATUS_NEEDS_MORE_INPUT;
+              break;
-            return SZ_OK;
           }
-          if (checkEndMarkNow && dummyRes != DUMMY_MATCH)
+          else if ((unsigned)dummyProcessed != processed)
+            break;
+
+          src += processed;
+          inSize -= processed;
+          (*srcLen) += processed;
+
+          if (res != SZ_OK)
           {
-            *status = LZMA_STATUS_NOT_FINISHED;
+            p->remainLen = kMatchSpecLen_Error_Data;
             return SZ_ERROR_DATA;
           }
         }
+        continue;
+      }
+
+      {
+        // we have some data in (p->tempBuf)
+        // in strict mode: tempBufSize is not enough for one Symbol decoding.
+        // in relaxed mode: tempBufSize not larger than required for one Symbol decoding.
+
+        unsigned rem = p->tempBufSize;
+        unsigned ahead = 0;
+        int dummyProcessed = -1;
+        
+        while (rem < LZMA_REQUIRED_INPUT_MAX && ahead < inSize)
+          p->tempBuf[rem++] = src[ahead++];
+        
+        // ahead - the size of new data copied from (src) to (p->tempBuf)
+        // rem   - the size of temp buffer including new data from (src)
+        
+        if (rem < LZMA_REQUIRED_INPUT_MAX || checkEndMarkNow)
+        {
+          const Byte *bufOut = p->tempBuf + rem;
+        
+          ELzmaDummy dummyRes = LzmaDec_TryDummy(p, p->tempBuf, &bufOut);
+          
+          if (dummyRes == DUMMY_INPUT_EOF)
+          {
+            if (rem >= LZMA_REQUIRED_INPUT_MAX)
+              break;
+            p->tempBufSize = rem;
+            (*srcLen) += (SizeT)ahead;
+            *status = LZMA_STATUS_NEEDS_MORE_INPUT;
+            return SZ_OK;
+          }
+          
+          dummyProcessed = (int)(bufOut - p->tempBuf);
+
+          if ((unsigned)dummyProcessed < p->tempBufSize)
+            break;
+
+          if (checkEndMarkNow && !IS_DUMMY_END_MARKER_POSSIBLE(dummyRes))
+          {
+            (*srcLen) += (unsigned)dummyProcessed - p->tempBufSize;
+            p->tempBufSize = (unsigned)dummyProcessed;
+            // p->remainLen = kMatchSpecLen_Error_Data;
+            RETURN__NOT_FINISHED__FOR_FINISH;
+          }
+        }
+
         p->buf = p->tempBuf;
-        if (LzmaDec_DecodeReal2(p, dicLimit, p->buf) != 0)
-          return SZ_ERROR_DATA;
         
         {
-          unsigned kkk = (unsigned)(p->buf - p->tempBuf);
+          // we decode one symbol from (p->tempBuf) here, so the (bufLimit) is equal to (p->buf)
-          if (rem < kkk)
+          int res = LzmaDec_DecodeReal2(p, dicLimit, p->buf);
-            return SZ_ERROR_FAIL; /* some internal error */
+
-          rem -= kkk;
+          SizeT processed = (SizeT)(p->buf - p->tempBuf);
-          if (lookAhead < rem)
+          rem = p->tempBufSize;
-            return SZ_ERROR_FAIL; /* some internal error */
+          
-          lookAhead -= rem;
+          if (dummyProcessed < 0)
+          {
+            if (processed > LZMA_REQUIRED_INPUT_MAX)
+              break;
+            if (processed < rem)
+              break;
+          }
+          else if ((unsigned)dummyProcessed != processed)
+            break;
+          
+          processed -= rem;
+
+          src += processed;
+          inSize -= processed;
+          (*srcLen) += processed;
+          p->tempBufSize = 0;
+          
+          if (res != SZ_OK)
+          {
+            p->remainLen = kMatchSpecLen_Error_Data;
+            return SZ_ERROR_DATA;
+          }
         }
-        (*srcLen) += (SizeT)lookAhead;
-        src += lookAhead;
-        inSize -= (SizeT)lookAhead;
-        p->tempBufSize = 0;
       }
+    }
   }
-  
+
-  if (p->code != 0)
+  /*  Some unexpected error: internal error of code, memory corruption or hardware failure */
-    return SZ_ERROR_DATA;
+  p->remainLen = kMatchSpecLen_Error_Fail;
-  *status = LZMA_STATUS_FINISHED_WITH_MARK;
+  return SZ_ERROR_FAIL;
-  return SZ_OK;
 }
 
 
+
 SRes LzmaDec_DecodeToBuf(CLzmaDec *p, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status)
 {
   SizeT outSize = *destLen;

libraries/lzma/C/LzmaDec.h

@@ -1,5 +1,5 @@
 /* LzmaDec.h -- LZMA Decoder
-2018-04-21 : Igor Pavlov : Public domain */
+2020-03-19 : Igor Pavlov : Public domain */
 
 #ifndef __LZMA_DEC_H
 #define __LZMA_DEC_H
@@ -181,6 +181,7 @@ Returns:
       LZMA_STATUS_NEEDS_MORE_INPUT
       LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK
   SZ_ERROR_DATA - Data error
+  SZ_ERROR_FAIL - Some unexpected error: internal error of code, memory corruption or hardware failure
 */
 
 SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit,
@@ -223,6 +224,7 @@ Returns:
   SZ_ERROR_MEM  - Memory allocation error
   SZ_ERROR_UNSUPPORTED - Unsupported properties
   SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src).
+  SZ_ERROR_FAIL - Some unexpected error: internal error of code, memory corruption or hardware failure
 */
 
 SRes LzmaDecode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen,

libraries/lzma/C/LzmaEnc.c

@@ -1,5 +1,5 @@
 /* LzmaEnc.c -- LZMA Encoder
-2019-01-10: Igor Pavlov : Public domain */
+2021-11-18: Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
@@ -12,6 +12,7 @@
 #include <stdio.h>
 #endif
 
+#include "CpuArch.h"
 #include "LzmaEnc.h"
 
 #include "LzFind.h"
@@ -19,12 +20,25 @@
 #include "LzFindMt.h"
 #endif
 
+/* the following LzmaEnc_* declarations is internal LZMA interface for LZMA2 encoder */
+
+SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, ISeqInStream *inStream, UInt32 keepWindowSize,
+    ISzAllocPtr alloc, ISzAllocPtr allocBig);
+SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen,
+    UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig);
+SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit,
+    Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize);
+const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp);
+void LzmaEnc_Finish(CLzmaEncHandle pp);
+void LzmaEnc_SaveState(CLzmaEncHandle pp);
+void LzmaEnc_RestoreState(CLzmaEncHandle pp);
+
 #ifdef SHOW_STAT
 static unsigned g_STAT_OFFSET = 0;
 #endif
 
-#define kLzmaMaxHistorySize ((UInt32)3 << 29)
+/* for good normalization speed we still reserve 256 MB before 4 GB range */
-/* #define kLzmaMaxHistorySize ((UInt32)7 << 29) */
+#define kLzmaMaxHistorySize ((UInt32)15 << 28)
 
 #define kNumTopBits 24
 #define kTopValue ((UInt32)1 << kNumTopBits)
@@ -36,7 +50,7 @@ static unsigned g_STAT_OFFSET = 0;
 
 #define kNumMoveReducingBits 4
 #define kNumBitPriceShiftBits 4
-#define kBitPrice (1 << kNumBitPriceShiftBits)
+// #define kBitPrice (1 << kNumBitPriceShiftBits)
 
 #define REP_LEN_COUNT 64
 
@@ -47,6 +61,7 @@ void LzmaEncProps_Init(CLzmaEncProps *p)
   p->reduceSize = (UInt64)(Int64)-1;
   p->lc = p->lp = p->pb = p->algo = p->fb = p->btMode = p->numHashBytes = p->numThreads = -1;
   p->writeEndMark = 0;
+  p->affinity = 0;
 }
 
 void LzmaEncProps_Normalize(CLzmaEncProps *p)
@@ -55,16 +70,21 @@ void LzmaEncProps_Normalize(CLzmaEncProps *p)
   if (level < 0) level = 5;
   p->level = level;
   
-  if (p->dictSize == 0) p->dictSize = (level <= 5 ? (1 << (level * 2 + 14)) : (level <= 7 ? (1 << 25) : (1 << 26)));
+  if (p->dictSize == 0)
+    p->dictSize =
+      ( level <= 3 ? ((UInt32)1 << (level * 2 + 16)) :
+      ( level <= 6 ? ((UInt32)1 << (level + 19)) :
+      ( level <= 7 ? ((UInt32)1 << 25) : ((UInt32)1 << 26)
+      )));
+
   if (p->dictSize > p->reduceSize)
   {
-    unsigned i;
+    UInt32 v = (UInt32)p->reduceSize;
-    UInt32 reduceSize = (UInt32)p->reduceSize;
+    const UInt32 kReduceMin = ((UInt32)1 << 12);
-    for (i = 11; i <= 30; i++)
+    if (v < kReduceMin)
-    {
+      v = kReduceMin;
-      if (reduceSize <= ((UInt32)2 << i)) { p->dictSize = ((UInt32)2 << i); break; }
+    if (p->dictSize > v)
-      if (reduceSize <= ((UInt32)3 << i)) { p->dictSize = ((UInt32)3 << i); break; }
+      p->dictSize = v;
-    }
   }
 
   if (p->lc < 0) p->lc = 3;
@@ -74,8 +94,8 @@ void LzmaEncProps_Normalize(CLzmaEncProps *p)
   if (p->algo < 0) p->algo = (level < 5 ? 0 : 1);
   if (p->fb < 0) p->fb = (level < 7 ? 32 : 64);
   if (p->btMode < 0) p->btMode = (p->algo == 0 ? 0 : 1);
-  if (p->numHashBytes < 0) p->numHashBytes = 4;
+  if (p->numHashBytes < 0) p->numHashBytes = (p->btMode ? 4 : 5);
-  if (p->mc == 0) p->mc = (16 + (p->fb >> 1)) >> (p->btMode ? 0 : 1);
+  if (p->mc == 0) p->mc = (16 + ((unsigned)p->fb >> 1)) >> (p->btMode ? 0 : 1);
   
   if (p->numThreads < 0)
     p->numThreads =
@@ -93,18 +113,85 @@ UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2)
   return props.dictSize;
 }
 
-#if (_MSC_VER >= 1400)
+
-/* BSR code is fast for some new CPUs */
+/*
-/* #define LZMA_LOG_BSR */
+x86/x64:
+
+BSR:
+  IF (SRC == 0) ZF = 1, DEST is undefined;
+                  AMD : DEST is unchanged;
+  IF (SRC != 0) ZF = 0; DEST is index of top non-zero bit
+  BSR is slow in some processors
+
+LZCNT:
+  IF (SRC  == 0) CF = 1, DEST is size_in_bits_of_register(src) (32 or 64)
+  IF (SRC  != 0) CF = 0, DEST = num_lead_zero_bits
+  IF (DEST == 0) ZF = 1;
+
+LZCNT works only in new processors starting from Haswell.
+if LZCNT is not supported by processor, then it's executed as BSR.
+LZCNT can be faster than BSR, if supported.
+*/
+
+// #define LZMA_LOG_BSR
+
+#if defined(MY_CPU_ARM_OR_ARM64) /* || defined(MY_CPU_X86_OR_AMD64) */
+
+  #if (defined(__clang__) && (__clang_major__ >= 6)) \
+      || (defined(__GNUC__) && (__GNUC__ >= 6))
+      #define LZMA_LOG_BSR
+  #elif defined(_MSC_VER) && (_MSC_VER >= 1300)
+    // #if defined(MY_CPU_ARM_OR_ARM64)
+      #define LZMA_LOG_BSR
+    // #endif
+  #endif
 #endif
 
+// #include <intrin.h>
+
 #ifdef LZMA_LOG_BSR
 
-#define kDicLogSizeMaxCompress 32
+#if defined(__clang__) \
+    || defined(__GNUC__)
 
-#define BSR2_RET(pos, res) { unsigned long zz; _BitScanReverse(&zz, (pos)); res = (zz + zz) + ((pos >> (zz - 1)) & 1); }
+/*
+  C code:                  : (30 - __builtin_clz(x))
+    gcc9/gcc10 for x64 /x86  : 30 - (bsr(x) xor 31)
+    clang10 for x64          : 31 + (bsr(x) xor -32)
+*/
 
-static unsigned GetPosSlot1(UInt32 pos)
+  #define MY_clz(x)  ((unsigned)__builtin_clz(x))
+  // __lzcnt32
+  // __builtin_ia32_lzcnt_u32
+
+#else  // #if defined(_MSC_VER)
+
+  #ifdef MY_CPU_ARM_OR_ARM64
+
+    #define MY_clz  _CountLeadingZeros
+
+  #else // if defined(MY_CPU_X86_OR_AMD64)
+
+    // #define MY_clz  __lzcnt  // we can use lzcnt (unsupported by old CPU)
+    // _BitScanReverse code is not optimal for some MSVC compilers
+    #define BSR2_RET(pos, res) { unsigned long zz; _BitScanReverse(&zz, (pos)); zz--; \
+      res = (zz + zz) + (pos >> zz); }
+
+  #endif // MY_CPU_X86_OR_AMD64
+
+#endif // _MSC_VER
+
+
+#ifndef BSR2_RET
+
+    #define BSR2_RET(pos, res) { unsigned zz = 30 - MY_clz(pos); \
+      res = (zz + zz) + (pos >> zz); }
+
+#endif
+
+
+unsigned GetPosSlot1(UInt32 pos);
+unsigned GetPosSlot1(UInt32 pos)
 {
   unsigned res;
   BSR2_RET(pos, res);
@@ -113,10 +200,10 @@ static unsigned GetPosSlot1(UInt32 pos)
 #define GetPosSlot2(pos, res) { BSR2_RET(pos, res); }
 #define GetPosSlot(pos, res) { if (pos < 2) res = pos; else BSR2_RET(pos, res); }
 
-#else
 
-#define kNumLogBits (9 + sizeof(size_t) / 2)
+#else // ! LZMA_LOG_BSR
-/* #define kNumLogBits (11 + sizeof(size_t) / 8 * 3) */
+
+#define kNumLogBits (11 + sizeof(size_t) / 8 * 3)
 
 #define kDicLogSizeMaxCompress ((kNumLogBits - 1) * 2 + 7)
 
@@ -163,7 +250,7 @@ static void LzmaEnc_FastPosInit(Byte *g_FastPos)
 #define GetPosSlot2(pos, res) { BSR2_RET(pos, res); }
 #define GetPosSlot(pos, res) { if (pos < kNumFullDistances) res = p->g_FastPos[pos & (kNumFullDistances - 1)]; else BSR2_RET(pos, res); }
 
-#endif
+#endif // LZMA_LOG_BSR
 
 
 #define LZMA_NUM_REPS 4
@@ -193,7 +280,7 @@ typedef struct
 
 #define kNumLenToPosStates 4
 #define kNumPosSlotBits 6
-#define kDicLogSizeMin 0
+// #define kDicLogSizeMin 0
 #define kDicLogSizeMax 32
 #define kDistTableSizeMax (kDicLogSizeMax * 2)
 
@@ -299,7 +386,7 @@ typedef UInt32 CProbPrice;
 typedef struct
 {
   void *matchFinderObj;
-  IMatchFinder matchFinder;
+  IMatchFinder2 matchFinder;
 
   unsigned optCur;
   unsigned optEnd;
@@ -344,10 +431,14 @@ typedef struct
   // begin of CMatchFinderMt is used in LZ thread
   CMatchFinderMt matchFinderMt;
   // end of CMatchFinderMt is used in BT and HASH threads
+  // #else
+  // CMatchFinder matchFinderBase;
   #endif
-
   CMatchFinder matchFinderBase;
 
+  
+  // we suppose that we have 8-bytes alignment after CMatchFinder
+ 
   #ifndef _7ZIP_ST
   Byte pad[128];
   #endif
@@ -355,8 +446,10 @@ typedef struct
   // LZ thread
   CProbPrice ProbPrices[kBitModelTotal >> kNumMoveReducingBits];
 
-  UInt32 matches[LZMA_MATCH_LEN_MAX * 2 + 2 + 1];
+  // we want {len , dist} pairs to be 8-bytes aligned in matches array
+  UInt32 matches[LZMA_MATCH_LEN_MAX * 2 + 2];
 
+  // we want 8-bytes alignment here
   UInt32 alignPrices[kAlignTableSize];
   UInt32 posSlotPrices[kNumLenToPosStates][kDistTableSizeMax];
   UInt32 distancesPrices[kNumLenToPosStates][kNumFullDistances];
@@ -385,12 +478,19 @@ typedef struct
 
   CSaveState saveState;
 
+  // BoolInt mf_Failure;
   #ifndef _7ZIP_ST
   Byte pad2[128];
   #endif
 } CLzmaEnc;
 
 
+#define MFB (p->matchFinderBase)
+/*
+#ifndef _7ZIP_ST
+#define MFB (p->matchFinderMt.MatchFinder)
+#endif
+*/
 
 #define COPY_ARR(dest, src, arr) memcpy(dest->arr, src->arr, sizeof(src->arr));
 
@@ -455,41 +555,51 @@ SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2)
 
   if (props.lc > LZMA_LC_MAX
       || props.lp > LZMA_LP_MAX
-      || props.pb > LZMA_PB_MAX
+      || props.pb > LZMA_PB_MAX)
-      || props.dictSize > ((UInt64)1 << kDicLogSizeMaxCompress)
-      || props.dictSize > kLzmaMaxHistorySize)
     return SZ_ERROR_PARAM;
 
+
+  if (props.dictSize > kLzmaMaxHistorySize)
+    props.dictSize = kLzmaMaxHistorySize;
+
+  #ifndef LZMA_LOG_BSR
+  {
+    const UInt64 dict64 = props.dictSize;
+    if (dict64 > ((UInt64)1 << kDicLogSizeMaxCompress))
+      return SZ_ERROR_PARAM;
+  }
+  #endif
+
   p->dictSize = props.dictSize;
   {
-    unsigned fb = props.fb;
+    unsigned fb = (unsigned)props.fb;
     if (fb < 5)
       fb = 5;
     if (fb > LZMA_MATCH_LEN_MAX)
       fb = LZMA_MATCH_LEN_MAX;
     p->numFastBytes = fb;
   }
-  p->lc = props.lc;
+  p->lc = (unsigned)props.lc;
-  p->lp = props.lp;
+  p->lp = (unsigned)props.lp;
-  p->pb = props.pb;
+  p->pb = (unsigned)props.pb;
   p->fastMode = (props.algo == 0);
   // p->_maxMode = True;
-  p->matchFinderBase.btMode = (Byte)(props.btMode ? 1 : 0);
+  MFB.btMode = (Byte)(props.btMode ? 1 : 0);
   {
     unsigned numHashBytes = 4;
     if (props.btMode)
     {
-      if (props.numHashBytes < 2)
+           if (props.numHashBytes <  2) numHashBytes = 2;
-        numHashBytes = 2;
+      else if (props.numHashBytes <  4) numHashBytes = (unsigned)props.numHashBytes;
-      else if (props.numHashBytes < 4)
-        numHashBytes = props.numHashBytes;
     }
-    p->matchFinderBase.numHashBytes = numHashBytes;
+    if (props.numHashBytes >= 5) numHashBytes = 5;
+
+    MFB.numHashBytes = numHashBytes;
   }
 
-  p->matchFinderBase.cutValue = props.mc;
+  MFB.cutValue = props.mc;
 
-  p->writeEndMark = props.writeEndMark;
+  p->writeEndMark = (BoolInt)props.writeEndMark;
 
   #ifndef _7ZIP_ST
   /*
@@ -500,6 +610,8 @@ SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2)
   }
   */
   p->multiThread = (props.numThreads > 1);
+  p->matchFinderMt.btSync.affinity =
+  p->matchFinderMt.hashSync.affinity = props.affinity;
   #endif
 
   return SZ_OK;
@@ -509,7 +621,7 @@ SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2)
 void LzmaEnc_SetDataSize(CLzmaEncHandle pp, UInt64 expectedDataSiize)
 {
   CLzmaEnc *p = (CLzmaEnc *)pp;
-  p->matchFinderBase.expectedDataSize = expectedDataSiize;
+  MFB.expectedDataSize = expectedDataSiize;
 }
 
 
@@ -536,8 +648,8 @@ static void RangeEnc_Construct(CRangeEnc *p)
   p->bufBase = NULL;
 }
 
-#define RangeEnc_GetProcessed(p)       ((p)->processed + ((p)->buf - (p)->bufBase) + (p)->cacheSize)
+#define RangeEnc_GetProcessed(p)       (        (p)->processed + (size_t)((p)->buf - (p)->bufBase) +         (p)->cacheSize)
-#define RangeEnc_GetProcessed_sizet(p) ((size_t)(p)->processed + ((p)->buf - (p)->bufBase) + (size_t)(p)->cacheSize)
+#define RangeEnc_GetProcessed_sizet(p) ((size_t)(p)->processed + (size_t)((p)->buf - (p)->bufBase) + (size_t)(p)->cacheSize)
 
 #define RC_BUF_SIZE (1 << 16)
 
@@ -556,12 +668,11 @@ static int RangeEnc_Alloc(CRangeEnc *p, ISzAllocPtr alloc)
 static void RangeEnc_Free(CRangeEnc *p, ISzAllocPtr alloc)
 {
   ISzAlloc_Free(alloc, p->bufBase);
-  p->bufBase = 0;
+  p->bufBase = NULL;
 }
 
 static void RangeEnc_Init(CRangeEnc *p)
 {
-  /* Stream.Init(); */
   p->range = 0xFFFFFFFF;
   p->cache = 0;
   p->low = 0;
@@ -575,12 +686,12 @@ static void RangeEnc_Init(CRangeEnc *p)
 
 MY_NO_INLINE static void RangeEnc_FlushStream(CRangeEnc *p)
 {
-  size_t num;
+  const size_t num = (size_t)(p->buf - p->bufBase);
-  if (p->res != SZ_OK)
+  if (p->res == SZ_OK)
-    return;
+  {
-  num = p->buf - p->bufBase;
+    if (num != ISeqOutStream_Write(p->outStream, p->bufBase, num))
-  if (num != ISeqOutStream_Write(p->outStream, p->bufBase, num))
+      p->res = SZ_ERROR_WRITE;
-    p->res = SZ_ERROR_WRITE;
+  }
   p->processed += num;
   p->buf = p->bufBase;
 }
@@ -656,7 +767,7 @@ static void RangeEnc_FlushData(CRangeEnc *p)
   range += newBound & mask; \
   mask &= (kBitModelTotal - ((1 << kNumMoveBits) - 1)); \
   mask += ((1 << kNumMoveBits) - 1); \
-  ttt += (Int32)(mask - ttt) >> kNumMoveBits; \
+  ttt += (UInt32)((Int32)(mask - ttt) >> kNumMoveBits); \
   *(prob) = (CLzmaProb)ttt; \
   RC_NORM(p) \
   }
@@ -749,7 +860,7 @@ static void LzmaEnc_InitPriceTables(CProbPrice *ProbPrices)
         bitCount++;
       }
     }
-    ProbPrices[i] = (CProbPrice)((kNumBitModelTotalBits << kCyclesBits) - 15 - bitCount);
+    ProbPrices[i] = (CProbPrice)(((unsigned)kNumBitModelTotalBits << kCyclesBits) - 15 - bitCount);
     // printf("\n%3d: %5d", i, ProbPrices[i]);
   }
 }
@@ -985,7 +1096,11 @@ static unsigned ReadMatchDistances(CLzmaEnc *p, unsigned *numPairsRes)
   
   p->additionalOffset++;
   p->numAvail = p->matchFinder.GetNumAvailableBytes(p->matchFinderObj);
-  numPairs = p->matchFinder.GetMatches(p->matchFinderObj, p->matches);
+  {
+    const UInt32 *d = p->matchFinder.GetMatches(p->matchFinderObj, p->matches);
+    // if (!d) { p->mf_Failure = True; *numPairsRes = 0;  return 0; }
+    numPairs = (unsigned)(d - p->matches);
+  }
   *numPairsRes = numPairs;
   
   #ifdef SHOW_STAT
@@ -1001,7 +1116,7 @@ static unsigned ReadMatchDistances(CLzmaEnc *p, unsigned *numPairsRes)
   if (numPairs == 0)
     return 0;
   {
-    unsigned len = p->matches[(size_t)numPairs - 2];
+    const unsigned len = p->matches[(size_t)numPairs - 2];
     if (len != p->numFastBytes)
       return len;
     {
@@ -1011,7 +1126,7 @@ static unsigned ReadMatchDistances(CLzmaEnc *p, unsigned *numPairsRes)
       {
         const Byte *p1 = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1;
         const Byte *p2 = p1 + len;
-        ptrdiff_t dif = (ptrdiff_t)-1 - p->matches[(size_t)numPairs - 1];
+        const ptrdiff_t dif = (ptrdiff_t)-1 - (ptrdiff_t)p->matches[(size_t)numPairs - 1];
         const Byte *lim = p1 + numAvail;
         for (; p2 != lim && *p2 == p2[dif]; p2++)
         {}
@@ -1167,6 +1282,8 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
       repLens[i] = len;
       if (len > repLens[repMaxIndex])
         repMaxIndex = i;
+      if (len == LZMA_MATCH_LEN_MAX) // 21.03 : optimization
+        break;
     }
     
     if (repLens[repMaxIndex] >= p->numFastBytes)
@@ -1179,10 +1296,12 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
     }
     
     matches = p->matches;
+    #define MATCHES  matches
+    // #define MATCHES  p->matches
     
     if (mainLen >= p->numFastBytes)
     {
-      p->backRes = matches[(size_t)numPairs - 1] + LZMA_NUM_REPS;
+      p->backRes = MATCHES[(size_t)numPairs - 1] + LZMA_NUM_REPS;
       MOVE_POS(p, mainLen - 1)
       return mainLen;
     }
@@ -1276,13 +1395,13 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
         if (len < 2)
           len = 2;
         else
-          while (len > matches[offs])
+          while (len > MATCHES[offs])
             offs += 2;
     
         for (; ; len++)
         {
           COptimal *opt;
-          UInt32 dist = matches[(size_t)offs + 1];
+          UInt32 dist = MATCHES[(size_t)offs + 1];
           UInt32 price = normalMatchPrice + GET_PRICE_LEN(&p->lenEnc, posState, len);
           unsigned lenToPosState = GetLenToPosState(len);
        
@@ -1306,7 +1425,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
             opt->extra = 0;
           }
           
-          if (len == matches[offs])
+          if (len == MATCHES[offs])
           {
             offs += 2;
             if (offs == numPairs)
@@ -1727,8 +1846,8 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
     if (newLen > numAvail)
     {
       newLen = numAvail;
-      for (numPairs = 0; newLen > matches[numPairs]; numPairs += 2);
+      for (numPairs = 0; newLen > MATCHES[numPairs]; numPairs += 2);
-      matches[numPairs] = (UInt32)newLen;
+      MATCHES[numPairs] = (UInt32)newLen;
       numPairs += 2;
     }
     
@@ -1747,9 +1866,9 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
       }
 
       offs = 0;
-      while (startLen > matches[offs])
+      while (startLen > MATCHES[offs])
         offs += 2;
-      dist = matches[(size_t)offs + 1];
+      dist = MATCHES[(size_t)offs + 1];
       
       // if (dist >= kNumFullDistances)
       GetPosSlot2(dist, posSlot);
@@ -1776,7 +1895,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
           }
         }
 
-        if (len == matches[offs])
+        if (len == MATCHES[offs])
         {
           // if (p->_maxMode) {
           // MATCH : LIT : REP_0
@@ -1841,7 +1960,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
           offs += 2;
           if (offs == numPairs)
             break;
-          dist = matches[(size_t)offs + 1];
+          dist = MATCHES[(size_t)offs + 1];
           // if (dist >= kNumFullDistances)
             GetPosSlot2(dist, posSlot);
         }
@@ -2059,8 +2178,23 @@ static SRes CheckErrors(CLzmaEnc *p)
     return p->result;
   if (p->rc.res != SZ_OK)
     p->result = SZ_ERROR_WRITE;
-  if (p->matchFinderBase.result != SZ_OK)
+
+  #ifndef _7ZIP_ST
+  if (
+      // p->mf_Failure ||
+        (p->mtMode &&
+          ( // p->matchFinderMt.failure_LZ_LZ ||
+            p->matchFinderMt.failure_LZ_BT))
+     )
+  {
+    p->result = MY_HRES_ERROR__INTERNAL_ERROR;
+    // printf("\nCheckErrors p->matchFinderMt.failureLZ\n");
+  }
+  #endif
+
+  if (MFB.result != SZ_OK)
     p->result = SZ_ERROR_READ;
+  
   if (p->result != SZ_OK)
     p->finished = True;
   return p->result;
@@ -2198,14 +2332,14 @@ MY_NO_INLINE static void FillDistancesPrices(CLzmaEnc *p)
 
 
 
-void LzmaEnc_Construct(CLzmaEnc *p)
+static void LzmaEnc_Construct(CLzmaEnc *p)
 {
   RangeEnc_Construct(&p->rc);
-  MatchFinder_Construct(&p->matchFinderBase);
+  MatchFinder_Construct(&MFB);
   
   #ifndef _7ZIP_ST
+  p->matchFinderMt.MatchFinder = &MFB;
   MatchFinderMt_Construct(&p->matchFinderMt);
-  p->matchFinderMt.MatchFinder = &p->matchFinderBase;
   #endif
 
   {
@@ -2221,7 +2355,6 @@ void LzmaEnc_Construct(CLzmaEnc *p)
   LzmaEnc_InitPriceTables(p->ProbPrices);
   p->litProbs = NULL;
   p->saveState.litProbs = NULL;
-
 }
 
 CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc)
@@ -2233,7 +2366,7 @@ CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc)
   return p;
 }
 
-void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAllocPtr alloc)
+static void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAllocPtr alloc)
 {
   ISzAlloc_Free(alloc, p->litProbs);
   ISzAlloc_Free(alloc, p->saveState.litProbs);
@@ -2241,13 +2374,13 @@ void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAllocPtr alloc)
   p->saveState.litProbs = NULL;
 }
 
-void LzmaEnc_Destruct(CLzmaEnc *p, ISzAllocPtr alloc, ISzAllocPtr allocBig)
+static void LzmaEnc_Destruct(CLzmaEnc *p, ISzAllocPtr alloc, ISzAllocPtr allocBig)
 {
   #ifndef _7ZIP_ST
   MatchFinderMt_Destruct(&p->matchFinderMt, allocBig);
   #endif
   
-  MatchFinder_Free(&p->matchFinderBase, allocBig);
+  MatchFinder_Free(&MFB, allocBig);
   LzmaEnc_FreeLits(p, alloc);
   RangeEnc_Free(&p->rc, alloc);
 }
@@ -2259,11 +2392,18 @@ void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig)
 }
 
 
+MY_NO_INLINE
 static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpackSize)
 {
   UInt32 nowPos32, startPos32;
   if (p->needInit)
   {
+    #ifndef _7ZIP_ST
+    if (p->mtMode)
+    {
+      RINOK(MatchFinderMt_InitMt(&p->matchFinderMt));
+    }
+    #endif
     p->matchFinder.Init(p->matchFinderObj);
     p->needInit = 0;
   }
@@ -2521,12 +2661,12 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa
           // { int y; for (y = 0; y < 100; y++) {
           FillDistancesPrices(p);
           // }}
-          LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices);
+          LenPriceEnc_UpdateTables(&p->lenEnc, (unsigned)1 << p->pb, &p->lenProbs, p->ProbPrices);
         }
         if (p->repLenEncCounter <= 0)
         {
           p->repLenEncCounter = REP_LEN_COUNT;
-          LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, &p->repLenProbs, p->ProbPrices);
+          LenPriceEnc_UpdateTables(&p->repLenEnc, (unsigned)1 << p->pb, &p->repLenProbs, p->ProbPrices);
         }
       }
     
@@ -2559,11 +2699,13 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa
 static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig)
 {
   UInt32 beforeSize = kNumOpts;
+  UInt32 dictSize;
+
   if (!RangeEnc_Alloc(&p->rc, alloc))
     return SZ_ERROR_MEM;
 
   #ifndef _7ZIP_ST
-  p->mtMode = (p->multiThread && !p->fastMode && (p->matchFinderBase.btMode != 0));
+  p->mtMode = (p->multiThread && !p->fastMode && (MFB.btMode != 0));
   #endif
 
   {
@@ -2582,36 +2724,56 @@ static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc,
     }
   }
 
-  p->matchFinderBase.bigHash = (Byte)(p->dictSize > kBigHashDicLimit ? 1 : 0);
+  MFB.bigHash = (Byte)(p->dictSize > kBigHashDicLimit ? 1 : 0);
 
-  if (beforeSize + p->dictSize < keepWindowSize)
+
-    beforeSize = keepWindowSize - p->dictSize;
+  dictSize = p->dictSize;
+  if (dictSize == ((UInt32)2 << 30) ||
+      dictSize == ((UInt32)3 << 30))
+  {
+    /* 21.03 : here we reduce the dictionary for 2 reasons:
+       1) we don't want 32-bit back_distance matches in decoder for 2 GB dictionary.
+       2) we want to elimate useless last MatchFinder_Normalize3() for corner cases,
+          where data size is aligned for 1 GB: 5/6/8 GB.
+          That reducing must be >= 1 for such corner cases. */
+    dictSize -= 1;
+  }
+
+  if (beforeSize + dictSize < keepWindowSize)
+    beforeSize = keepWindowSize - dictSize;
+
+  /* in worst case we can look ahead for
+        max(LZMA_MATCH_LEN_MAX, numFastBytes + 1 + numFastBytes) bytes.
+     we send larger value for (keepAfter) to MantchFinder_Create():
+        (numFastBytes + LZMA_MATCH_LEN_MAX + 1)
+  */
 
   #ifndef _7ZIP_ST
   if (p->mtMode)
   {
-    RINOK(MatchFinderMt_Create(&p->matchFinderMt, p->dictSize, beforeSize, p->numFastBytes,
+    RINOK(MatchFinderMt_Create(&p->matchFinderMt, dictSize, beforeSize,
-        LZMA_MATCH_LEN_MAX
+        p->numFastBytes, LZMA_MATCH_LEN_MAX + 1 /* 18.04 */
-        + 1  /* 18.04 */
         , allocBig));
     p->matchFinderObj = &p->matchFinderMt;
-    p->matchFinderBase.bigHash = (Byte)(
+    MFB.bigHash = (Byte)(
-        (p->dictSize > kBigHashDicLimit && p->matchFinderBase.hashMask >= 0xFFFFFF) ? 1 : 0);
+        (p->dictSize > kBigHashDicLimit && MFB.hashMask >= 0xFFFFFF) ? 1 : 0);
     MatchFinderMt_CreateVTable(&p->matchFinderMt, &p->matchFinder);
   }
   else
   #endif
   {
-    if (!MatchFinder_Create(&p->matchFinderBase, p->dictSize, beforeSize, p->numFastBytes, LZMA_MATCH_LEN_MAX, allocBig))
+    if (!MatchFinder_Create(&MFB, dictSize, beforeSize,
+        p->numFastBytes, LZMA_MATCH_LEN_MAX + 1 /* 21.03 */
+        , allocBig))
       return SZ_ERROR_MEM;
-    p->matchFinderObj = &p->matchFinderBase;
+    p->matchFinderObj = &MFB;
-    MatchFinder_CreateVTable(&p->matchFinderBase, &p->matchFinder);
+    MatchFinder_CreateVTable(&MFB, &p->matchFinder);
   }
   
   return SZ_OK;
 }
 
-void LzmaEnc_Init(CLzmaEnc *p)
+static void LzmaEnc_Init(CLzmaEnc *p)
 {
   unsigned i;
   p->state = 0;
@@ -2675,12 +2837,14 @@ void LzmaEnc_Init(CLzmaEnc *p)
 
   p->additionalOffset = 0;
 
-  p->pbMask = (1 << p->pb) - 1;
+  p->pbMask = ((unsigned)1 << p->pb) - 1;
   p->lpMask = ((UInt32)0x100 << p->lp) - ((unsigned)0x100 >> p->lc);
+
+  // p->mf_Failure = False;
 }
 
 
-void LzmaEnc_InitPrices(CLzmaEnc *p)
+static void LzmaEnc_InitPrices(CLzmaEnc *p)
 {
   if (!p->fastMode)
   {
@@ -2694,8 +2858,8 @@ void LzmaEnc_InitPrices(CLzmaEnc *p)
 
   p->repLenEncCounter = REP_LEN_COUNT;
 
-  LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices);
+  LenPriceEnc_UpdateTables(&p->lenEnc, (unsigned)1 << p->pb, &p->lenProbs, p->ProbPrices);
-  LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, &p->repLenProbs, p->ProbPrices);
+  LenPriceEnc_UpdateTables(&p->repLenEnc, (unsigned)1 << p->pb, &p->repLenProbs, p->ProbPrices);
 }
 
 static SRes LzmaEnc_AllocAndInit(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig)
@@ -2719,7 +2883,7 @@ static SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInS
     ISzAllocPtr alloc, ISzAllocPtr allocBig)
 {
   CLzmaEnc *p = (CLzmaEnc *)pp;
-  p->matchFinderBase.stream = inStream;
+  MFB.stream = inStream;
   p->needInit = 1;
   p->rc.outStream = outStream;
   return LzmaEnc_AllocAndInit(p, 0, alloc, allocBig);
@@ -2730,16 +2894,16 @@ SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp,
     ISzAllocPtr alloc, ISzAllocPtr allocBig)
 {
   CLzmaEnc *p = (CLzmaEnc *)pp;
-  p->matchFinderBase.stream = inStream;
+  MFB.stream = inStream;
   p->needInit = 1;
   return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig);
 }
 
 static void LzmaEnc_SetInputBuf(CLzmaEnc *p, const Byte *src, SizeT srcLen)
 {
-  p->matchFinderBase.directInput = 1;
+  MFB.directInput = 1;
-  p->matchFinderBase.bufferBase = (Byte *)src;
+  MFB.bufferBase = (Byte *)src;
-  p->matchFinderBase.directInputRem = srcLen;
+  MFB.directInputRem = srcLen;
 }
 
 SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen,
@@ -2781,19 +2945,23 @@ static size_t SeqOutStreamBuf_Write(const ISeqOutStream *pp, const void *data, s
     size = p->rem;
     p->overflow = True;
   }
-  memcpy(p->data, data, size);
+  if (size != 0)
-  p->rem -= size;
+  {
-  p->data += size;
+    memcpy(p->data, data, size);
+    p->rem -= size;
+    p->data += size;
+  }
   return size;
 }
 
 
+/*
 UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp)
 {
   const CLzmaEnc *p = (CLzmaEnc *)pp;
   return p->matchFinder.GetNumAvailableBytes(p->matchFinderObj);
 }
-
+*/
 
 const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp)
 {
@@ -2841,6 +3009,7 @@ SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit,
 }
 
 
+MY_NO_INLINE
 static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress)
 {
   SRes res = SZ_OK;
@@ -2870,7 +3039,7 @@ static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress)
   LzmaEnc_Finish(p);
 
   /*
-  if (res == SZ_OK && !Inline_MatchFinder_IsFinishedOK(&p->matchFinderBase))
+  if (res == SZ_OK && !Inline_MatchFinder_IsFinishedOK(&MFB))
     res = SZ_ERROR_FAIL;
   }
   */
@@ -2889,35 +3058,43 @@ SRes LzmaEnc_Encode(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *i
 
 SRes LzmaEnc_WriteProperties(CLzmaEncHandle pp, Byte *props, SizeT *size)
 {
-  CLzmaEnc *p = (CLzmaEnc *)pp;
-  unsigned i;
-  UInt32 dictSize = p->dictSize;
   if (*size < LZMA_PROPS_SIZE)
     return SZ_ERROR_PARAM;
   *size = LZMA_PROPS_SIZE;
-  props[0] = (Byte)((p->pb * 5 + p->lp) * 9 + p->lc);
-
-  if (dictSize >= ((UInt32)1 << 22))
   {
-    UInt32 kDictMask = ((UInt32)1 << 20) - 1;
+    const CLzmaEnc *p = (const CLzmaEnc *)pp;
-    if (dictSize < (UInt32)0xFFFFFFFF - kDictMask)
+    const UInt32 dictSize = p->dictSize;
-      dictSize = (dictSize + kDictMask) & ~kDictMask;
+    UInt32 v;
-  }
+    props[0] = (Byte)((p->pb * 5 + p->lp) * 9 + p->lc);
-  else for (i = 11; i <= 30; i++)
+    
-  {
+    // we write aligned dictionary value to properties for lzma decoder
-    if (dictSize <= ((UInt32)2 << i)) { dictSize = (2 << i); break; }
+    if (dictSize >= ((UInt32)1 << 21))
-    if (dictSize <= ((UInt32)3 << i)) { dictSize = (3 << i); break; }
+    {
-  }
+      const UInt32 kDictMask = ((UInt32)1 << 20) - 1;
+      v = (dictSize + kDictMask) & ~kDictMask;
+      if (v < dictSize)
+        v = dictSize;
+    }
+    else
+    {
+      unsigned i = 11 * 2;
+      do
+      {
+        v = (UInt32)(2 + (i & 1)) << (i >> 1);
+        i++;
+      }
+      while (v < dictSize);
+    }
 
-  for (i = 0; i < 4; i++)
+    SetUi32(props + 1, v);
-    props[1 + i] = (Byte)(dictSize >> (8 * i));
+    return SZ_OK;
-  return SZ_OK;
+  }
 }
 
 
 unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle pp)
 {
-  return ((CLzmaEnc *)pp)->writeEndMark;
+  return (unsigned)((CLzmaEnc *)pp)->writeEndMark;
 }
 
 
@@ -2974,3 +3151,15 @@ SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen,
   LzmaEnc_Destroy(p, alloc, allocBig);
   return res;
 }
+
+
+/*
+#ifndef _7ZIP_ST
+void LzmaEnc_GetLzThreads(CLzmaEncHandle pp, HANDLE lz_threads[2])
+{
+  const CLzmaEnc *p = (CLzmaEnc *)pp;
+  lz_threads[0] = p->matchFinderMt.hashSync.thread;
+  lz_threads[1] = p->matchFinderMt.btSync.thread;
+}
+#endif
+*/

libraries/lzma/C/LzmaEnc.h

@@ -1,5 +1,5 @@
 /*  LzmaEnc.h -- LZMA Encoder
-2017-07-27 : Igor Pavlov : Public domain */
+2019-10-30 : Igor Pavlov : Public domain */
 
 #ifndef __LZMA_ENC_H
 #define __LZMA_ENC_H
@@ -29,6 +29,8 @@ typedef struct _CLzmaEncProps
 
   UInt64 reduceSize; /* estimated size of data that will be compressed. default = (UInt64)(Int64)-1.
                         Encoder uses this value to reduce dictionary size */
+
+  UInt64 affinity;
 } CLzmaEncProps;
 
 void LzmaEncProps_Init(CLzmaEncProps *p);

libraries/lzma/C/Ppmd.h

@@ -1,5 +1,5 @@
 /* Ppmd.h -- PPMD codec common code
-2017-04-03 : Igor Pavlov : Public domain
+2021-04-13 : Igor Pavlov : Public domain
 This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */
 
 #ifndef __PPMD_H
@@ -9,7 +9,16 @@ This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */
 
 EXTERN_C_BEGIN
 
-#ifdef MY_CPU_32BIT
+#if defined(MY_CPU_SIZEOF_POINTER) && (MY_CPU_SIZEOF_POINTER == 4)
+/*
+   PPMD code always uses 32-bit internal fields in PPMD structures to store internal references in main block.
+   if (PPMD_32BIT is     defined), the PPMD code stores internal pointers to 32-bit reference fields.
+   if (PPMD_32BIT is NOT defined), the PPMD code stores internal UInt32 offsets to reference fields.
+   if (pointer size is 64-bit), then (PPMD_32BIT) mode is not allowed,
+   if (pointer size is 32-bit), then (PPMD_32BIT) mode is optional,
+     and it's allowed to disable PPMD_32BIT mode even if pointer is 32-bit.
+   PPMD code works slightly faster in (PPMD_32BIT) mode.
+*/
   #define PPMD_32BIT
 #endif
 
@@ -28,7 +37,7 @@ EXTERN_C_BEGIN
 #define PPMD_N4 ((128 + 3 - 1 * PPMD_N1 - 2 * PPMD_N2 - 3 * PPMD_N3) / 4)
 #define PPMD_NUM_INDEXES (PPMD_N1 + PPMD_N2 + PPMD_N3 + PPMD_N4)
 
-#pragma pack(push, 1)
+MY_CPU_pragma_pack_push_1
 /* Most compilers works OK here even without #pragma pack(push, 1), but some GCC compilers need it. */
 
 /* SEE-contexts for PPM-contexts with masked symbols */
@@ -40,41 +49,114 @@ typedef struct
 } CPpmd_See;
 
 #define Ppmd_See_Update(p)  if ((p)->Shift < PPMD_PERIOD_BITS && --(p)->Count == 0) \
-    { (p)->Summ <<= 1; (p)->Count = (Byte)(3 << (p)->Shift++); }
+    { (p)->Summ = (UInt16)((p)->Summ << 1); (p)->Count = (Byte)(3 << (p)->Shift++); }
+
 
 typedef struct
 {
   Byte Symbol;
   Byte Freq;
-  UInt16 SuccessorLow;
+  UInt16 Successor_0;
-  UInt16 SuccessorHigh;
+  UInt16 Successor_1;
 } CPpmd_State;
 
-#pragma pack(pop)
+typedef struct CPpmd_State2_
+{
+  Byte Symbol;
+  Byte Freq;
+} CPpmd_State2;
 
-typedef
+typedef struct CPpmd_State4_
-  #ifdef PPMD_32BIT
+{
-    CPpmd_State *
+  UInt16 Successor_0;
-  #else
+  UInt16 Successor_1;
-    UInt32
+} CPpmd_State4;
-  #endif
-  CPpmd_State_Ref;
 
-typedef
+MY_CPU_pragma_pop
-  #ifdef PPMD_32BIT
+
-    void *
+/*
-  #else
+   PPMD code can write full CPpmd_State structure data to CPpmd*_Context
-    UInt32
+      at (byte offset = 2) instead of some fields of original CPpmd*_Context structure.
-  #endif
+   
-  CPpmd_Void_Ref;
+   If we use pointers to different types, but that point to shared
+   memory space, we can have aliasing problem (strict aliasing).
+   
+   XLC compiler in -O2 mode can change the order of memory write instructions
+   in relation to read instructions, if we have use pointers to different types.
+   
+   To solve that aliasing problem we use combined CPpmd*_Context structure
+   with unions that contain the fields from both structures:
+   the original CPpmd*_Context and CPpmd_State.
+   So we can access the fields from both structures via one pointer,
+   and the compiler doesn't change the order of write instructions
+   in relation to read instructions.
+
+   If we don't use memory write instructions to shared memory in
+   some local code, and we use only reading instructions (read only),
+   then probably it's safe to use pointers to different types for reading.
+*/
+  
+
+
+#ifdef PPMD_32BIT
+
+  #define Ppmd_Ref_Type(type)   type *
+  #define Ppmd_GetRef(p, ptr)   (ptr)
+  #define Ppmd_GetPtr(p, ptr)   (ptr)
+  #define Ppmd_GetPtr_Type(p, ptr, note_type) (ptr)
+
+#else
+
+  #define Ppmd_Ref_Type(type)   UInt32
+  #define Ppmd_GetRef(p, ptr)   ((UInt32)((Byte *)(ptr) - (p)->Base))
+  #define Ppmd_GetPtr(p, offs)  ((void *)((p)->Base + (offs)))
+  #define Ppmd_GetPtr_Type(p, offs, type) ((type *)Ppmd_GetPtr(p, offs))
+
+#endif // PPMD_32BIT
+
+
+typedef Ppmd_Ref_Type(CPpmd_State) CPpmd_State_Ref;
+typedef Ppmd_Ref_Type(void)        CPpmd_Void_Ref;
+typedef Ppmd_Ref_Type(Byte)        CPpmd_Byte_Ref;
+
+
+/*
+#ifdef MY_CPU_LE_UNALIGN
+// the unaligned 32-bit access latency can be too large, if the data is not in L1 cache.
+#define Ppmd_GET_SUCCESSOR(p) ((CPpmd_Void_Ref)*(const UInt32 *)(const void *)&(p)->Successor_0)
+#define Ppmd_SET_SUCCESSOR(p, v) *(UInt32 *)(void *)(void *)&(p)->Successor_0 = (UInt32)(v)
+
+#else
+*/
+
+/*
+   We can write 16-bit halves to 32-bit (Successor) field in any selected order.
+   But the native order is more consistent way.
+   So we use the native order, if LE/BE order can be detected here at compile time.
+*/
+
+#ifdef MY_CPU_BE
+
+  #define Ppmd_GET_SUCCESSOR(p) \
+    ( (CPpmd_Void_Ref) (((UInt32)(p)->Successor_0 << 16) | (p)->Successor_1) )
+
+  #define Ppmd_SET_SUCCESSOR(p, v) { \
+    (p)->Successor_0 = (UInt16)(((UInt32)(v) >> 16) /* & 0xFFFF */); \
+    (p)->Successor_1 = (UInt16)((UInt32)(v) /* & 0xFFFF */); }
+
+#else
+
+  #define Ppmd_GET_SUCCESSOR(p) \
+    ( (CPpmd_Void_Ref) ((p)->Successor_0 | ((UInt32)(p)->Successor_1 << 16)) )
+
+  #define Ppmd_SET_SUCCESSOR(p, v) { \
+    (p)->Successor_0 = (UInt16)((UInt32)(v) /* & 0xFFFF */); \
+    (p)->Successor_1 = (UInt16)(((UInt32)(v) >> 16) /* & 0xFFFF */); }
+
+#endif
+
+// #endif
 
-typedef
-  #ifdef PPMD_32BIT
-    Byte *
-  #else
-    UInt32
-  #endif
-  CPpmd_Byte_Ref;
 
 #define PPMD_SetAllBitsIn256Bytes(p) \
   { size_t z; for (z = 0; z < 256 / sizeof(p[0]); z += 8) { \

libraries/lzma/C/Ppmd7.h

@@ -1,10 +1,8 @@
-/* Ppmd7.h -- PPMdH compression codec
+/* Ppmd7.h -- Ppmd7 (PPMdH) compression codec
-2018-07-04 : Igor Pavlov : Public domain
+2021-04-13 : Igor Pavlov : Public domain
-This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */
+This code is based on:
-
+  PPMd var.H (2001): Dmitry Shkarin : Public domain */
-/* This code supports virtual RangeDecoder and includes the implementation
+ 
-of RangeCoder from 7z, instead of RangeCoder from original PPMd var.H.
-If you need the compatibility with original PPMd var.H, you can use external RangeDecoder */
 
 #ifndef __PPMD7_H
 #define __PPMD7_H
@@ -21,23 +19,56 @@ EXTERN_C_BEGIN
 
 struct CPpmd7_Context_;
 
-typedef
+typedef Ppmd_Ref_Type(struct CPpmd7_Context_) CPpmd7_Context_Ref;
-  #ifdef PPMD_32BIT
+
-    struct CPpmd7_Context_ *
+// MY_CPU_pragma_pack_push_1
-  #else
-    UInt32
-  #endif
-  CPpmd7_Context_Ref;
 
 typedef struct CPpmd7_Context_
 {
   UInt16 NumStats;
-  UInt16 SummFreq;
+
-  CPpmd_State_Ref Stats;
+
+  union
+  {
+    UInt16 SummFreq;
+    CPpmd_State2 State2;
+  } Union2;
+
+  union
+  {
+    CPpmd_State_Ref Stats;
+    CPpmd_State4 State4;
+  } Union4;
+
   CPpmd7_Context_Ref Suffix;
 } CPpmd7_Context;
 
-#define Ppmd7Context_OneState(p) ((CPpmd_State *)&(p)->SummFreq)
+// MY_CPU_pragma_pop
+
+#define Ppmd7Context_OneState(p) ((CPpmd_State *)&(p)->Union2)
+
+
+
+
+typedef struct
+{
+  UInt32 Range;
+  UInt32 Code;
+  UInt32 Low;
+  IByteIn *Stream;
+} CPpmd7_RangeDec;
+
+
+typedef struct
+{
+  UInt32 Range;
+  Byte Cache;
+  // Byte _dummy_[3];
+  UInt64 Low;
+  UInt64 CacheSize;
+  IByteOut *Stream;
+} CPpmd7z_RangeEnc;
+
 
 typedef struct
 {
@@ -48,17 +79,30 @@ typedef struct
 
   UInt32 Size;
   UInt32 GlueCount;
-  Byte *Base, *LoUnit, *HiUnit, *Text, *UnitsStart;
   UInt32 AlignOffset;
+  Byte *Base, *LoUnit, *HiUnit, *Text, *UnitsStart;
 
-  Byte Indx2Units[PPMD_NUM_INDEXES];
+
+  
+  
+  union
+  {
+    CPpmd7_RangeDec dec;
+    CPpmd7z_RangeEnc enc;
+  } rc;
+  
+  Byte Indx2Units[PPMD_NUM_INDEXES + 2]; // +2 for alignment
   Byte Units2Indx[128];
   CPpmd_Void_Ref FreeList[PPMD_NUM_INDEXES];
-  Byte NS2Indx[256], NS2BSIndx[256], HB2Flag[256];
+
+  Byte NS2BSIndx[256], NS2Indx[256];
+  Byte ExpEscape[16];
   CPpmd_See DummySee, See[25][16];
   UInt16 BinSumm[128][64];
+  // int LastSymbol;
 } CPpmd7;
 
+
 void Ppmd7_Construct(CPpmd7 *p);
 BoolInt Ppmd7_Alloc(CPpmd7 *p, UInt32 size, ISzAllocPtr alloc);
 void Ppmd7_Free(CPpmd7 *p, ISzAllocPtr alloc);
@@ -68,74 +112,69 @@ void Ppmd7_Init(CPpmd7 *p, unsigned maxOrder);
 
 /* ---------- Internal Functions ---------- */
 
-extern const Byte PPMD7_kExpEscape[16];
+#define Ppmd7_GetPtr(p, ptr)     Ppmd_GetPtr(p, ptr)
-
+#define Ppmd7_GetContext(p, ptr) Ppmd_GetPtr_Type(p, ptr, CPpmd7_Context)
-#ifdef PPMD_32BIT
+#define Ppmd7_GetStats(p, ctx)   Ppmd_GetPtr_Type(p, (ctx)->Union4.Stats, CPpmd_State)
-  #define Ppmd7_GetPtr(p, ptr) (ptr)
-  #define Ppmd7_GetContext(p, ptr) (ptr)
-  #define Ppmd7_GetStats(p, ctx) ((ctx)->Stats)
-#else
-  #define Ppmd7_GetPtr(p, offs) ((void *)((p)->Base + (offs)))
-  #define Ppmd7_GetContext(p, offs) ((CPpmd7_Context *)Ppmd7_GetPtr((p), (offs)))
-  #define Ppmd7_GetStats(p, ctx) ((CPpmd_State *)Ppmd7_GetPtr((p), ((ctx)->Stats)))
-#endif
 
 void Ppmd7_Update1(CPpmd7 *p);
 void Ppmd7_Update1_0(CPpmd7 *p);
 void Ppmd7_Update2(CPpmd7 *p);
-void Ppmd7_UpdateBin(CPpmd7 *p);
+
+#define PPMD7_HiBitsFlag_3(sym) ((((unsigned)sym + 0xC0) >> (8 - 3)) & (1 << 3))
+#define PPMD7_HiBitsFlag_4(sym) ((((unsigned)sym + 0xC0) >> (8 - 4)) & (1 << 4))
+// #define PPMD7_HiBitsFlag_3(sym) ((sym) < 0x40 ? 0 : (1 << 3))
+// #define PPMD7_HiBitsFlag_4(sym) ((sym) < 0x40 ? 0 : (1 << 4))
 
 #define Ppmd7_GetBinSumm(p) \
-    &p->BinSumm[(size_t)(unsigned)Ppmd7Context_OneState(p->MinContext)->Freq - 1][p->PrevSuccess + \
+    &p->BinSumm[(size_t)(unsigned)Ppmd7Context_OneState(p->MinContext)->Freq - 1] \
-    p->NS2BSIndx[(size_t)Ppmd7_GetContext(p, p->MinContext->Suffix)->NumStats - 1] + \
+    [ p->PrevSuccess + ((p->RunLength >> 26) & 0x20) \
-    (p->HiBitsFlag = p->HB2Flag[p->FoundState->Symbol]) + \
+    + p->NS2BSIndx[(size_t)Ppmd7_GetContext(p, p->MinContext->Suffix)->NumStats - 1] \
-    2 * p->HB2Flag[(unsigned)Ppmd7Context_OneState(p->MinContext)->Symbol] + \
+    + PPMD7_HiBitsFlag_4(Ppmd7Context_OneState(p->MinContext)->Symbol) \
-    ((p->RunLength >> 26) & 0x20)]
+    + (p->HiBitsFlag = PPMD7_HiBitsFlag_3(p->FoundState->Symbol)) ]
 
 CPpmd_See *Ppmd7_MakeEscFreq(CPpmd7 *p, unsigned numMasked, UInt32 *scale);
 
 
+/*
+We support two versions of Ppmd7 (PPMdH) methods that use same CPpmd7 structure:
+  1) Ppmd7a_*: original PPMdH
+  2) Ppmd7z_*: modified PPMdH with 7z Range Coder
+Ppmd7_*: the structures and functions that are common for both versions of PPMd7 (PPMdH)
+*/
+
 /* ---------- Decode ---------- */
 
-typedef struct IPpmd7_RangeDec IPpmd7_RangeDec;
+#define PPMD7_SYM_END    (-1)
+#define PPMD7_SYM_ERROR  (-2)
 
-struct IPpmd7_RangeDec
+/*
-{
+You must set (CPpmd7::rc.dec.Stream) before Ppmd7*_RangeDec_Init()
-  UInt32 (*GetThreshold)(const IPpmd7_RangeDec *p, UInt32 total);
-  void (*Decode)(const IPpmd7_RangeDec *p, UInt32 start, UInt32 size);
-  UInt32 (*DecodeBit)(const IPpmd7_RangeDec *p, UInt32 size0);
-};
 
-typedef struct
+Ppmd7*_DecodeSymbol()
-{
+out:
-  IPpmd7_RangeDec vt;
+  >= 0 : decoded byte
-  UInt32 Range;
+    -1 : PPMD7_SYM_END   : End of payload marker
-  UInt32 Code;
+    -2 : PPMD7_SYM_ERROR : Data error
-  IByteIn *Stream;
+*/
-} CPpmd7z_RangeDec;
 
-void Ppmd7z_RangeDec_CreateVTable(CPpmd7z_RangeDec *p);
+/* Ppmd7a_* : original PPMdH */
-BoolInt Ppmd7z_RangeDec_Init(CPpmd7z_RangeDec *p);
+BoolInt Ppmd7a_RangeDec_Init(CPpmd7_RangeDec *p);
+#define Ppmd7a_RangeDec_IsFinishedOK(p) ((p)->Code == 0)
+int Ppmd7a_DecodeSymbol(CPpmd7 *p);
+
+/* Ppmd7z_* : modified PPMdH with 7z Range Coder */
+BoolInt Ppmd7z_RangeDec_Init(CPpmd7_RangeDec *p);
 #define Ppmd7z_RangeDec_IsFinishedOK(p) ((p)->Code == 0)
-
+int Ppmd7z_DecodeSymbol(CPpmd7 *p);
-int Ppmd7_DecodeSymbol(CPpmd7 *p, const IPpmd7_RangeDec *rc);
+// Byte *Ppmd7z_DecodeSymbols(CPpmd7 *p, Byte *buf, const Byte *lim);
 
 
 /* ---------- Encode ---------- */
 
-typedef struct
+void Ppmd7z_Init_RangeEnc(CPpmd7 *p);
-{
+void Ppmd7z_Flush_RangeEnc(CPpmd7 *p);
-  UInt64 Low;
+// void Ppmd7z_EncodeSymbol(CPpmd7 *p, int symbol);
-  UInt32 Range;
+void Ppmd7z_EncodeSymbols(CPpmd7 *p, const Byte *buf, const Byte *lim);
-  Byte Cache;
-  UInt64 CacheSize;
-  IByteOut *Stream;
-} CPpmd7z_RangeEnc;
-
-void Ppmd7z_RangeEnc_Init(CPpmd7z_RangeEnc *p);
-void Ppmd7z_RangeEnc_FlushData(CPpmd7z_RangeEnc *p);
-
-void Ppmd7_EncodeSymbol(CPpmd7 *p, CPpmd7z_RangeEnc *rc, int symbol);
 
 EXTERN_C_END
  

libraries/lzma/C/Ppmd7Dec.c

@@ -1,6 +1,8 @@
-/* Ppmd7Dec.c -- PPMdH Decoder
+/* Ppmd7Dec.c -- Ppmd7z (PPMdH with 7z Range Coder) Decoder
-2018-07-04 : Igor Pavlov : Public domain
+2021-04-13 : Igor Pavlov : Public domain
-This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */
+This code is based on:
+  PPMd var.H (2001): Dmitry Shkarin : Public domain */
+
 
 #include "Precomp.h"
 
@@ -8,184 +10,288 @@ This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */
 
 #define kTopValue (1 << 24)
 
-BoolInt Ppmd7z_RangeDec_Init(CPpmd7z_RangeDec *p)
+
+#define READ_BYTE(p) IByteIn_Read((p)->Stream)
+
+BoolInt Ppmd7z_RangeDec_Init(CPpmd7_RangeDec *p)
 {
   unsigned i;
   p->Code = 0;
   p->Range = 0xFFFFFFFF;
-  if (IByteIn_Read(p->Stream) != 0)
+  if (READ_BYTE(p) != 0)
     return False;
   for (i = 0; i < 4; i++)
-    p->Code = (p->Code << 8) | IByteIn_Read(p->Stream);
+    p->Code = (p->Code << 8) | READ_BYTE(p);
   return (p->Code < 0xFFFFFFFF);
 }
 
-#define GET_Ppmd7z_RangeDec CPpmd7z_RangeDec *p = CONTAINER_FROM_VTBL(pp, CPpmd7z_RangeDec, vt);
+#define RC_NORM_BASE(p) if ((p)->Range < kTopValue) \
- 
+  { (p)->Code = ((p)->Code << 8) | READ_BYTE(p); (p)->Range <<= 8;
-static UInt32 Range_GetThreshold(const IPpmd7_RangeDec *pp, UInt32 total)
+
+#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 R (&p->rc.dec)
+
+MY_FORCE_INLINE
+// MY_NO_INLINE
+static void RangeDec_Decode(CPpmd7 *p, UInt32 start, UInt32 size)
 {
-  GET_Ppmd7z_RangeDec
+
-  return p->Code / (p->Range /= total);
+  
+  R->Code -= start * R->Range;
+  R->Range *= size;
+  RC_NORM_LOCAL(R)
 }
 
-static void Range_Normalize(CPpmd7z_RangeDec *p)
+#define RC_Decode(start, size) RangeDec_Decode(p, start, size);
-{
+#define RC_DecodeFinal(start, size) RC_Decode(start, size) RC_NORM_REMOTE(R)
-  if (p->Range < kTopValue)
+#define RC_GetThreshold(total) (R->Code / (R->Range /= (total)))
-  {
-    p->Code = (p->Code << 8) | IByteIn_Read(p->Stream);
-    p->Range <<= 8;
-    if (p->Range < kTopValue)
-    {
-      p->Code = (p->Code << 8) | IByteIn_Read(p->Stream);
-      p->Range <<= 8;
-    }
-  }
-}
-
-static void Range_Decode(const IPpmd7_RangeDec *pp, UInt32 start, UInt32 size)
-{
-  GET_Ppmd7z_RangeDec
-  p->Code -= start * p->Range;
-  p->Range *= size;
-  Range_Normalize(p);
-}
-
-static UInt32 Range_DecodeBit(const IPpmd7_RangeDec *pp, UInt32 size0)
-{
-  GET_Ppmd7z_RangeDec
-  UInt32 newBound = (p->Range >> 14) * size0;
-  UInt32 symbol;
-  if (p->Code < newBound)
-  {
-    symbol = 0;
-    p->Range = newBound;
-  }
-  else
-  {
-    symbol = 1;
-    p->Code -= newBound;
-    p->Range -= newBound;
-  }
-  Range_Normalize(p);
-  return symbol;
-}
-
-void Ppmd7z_RangeDec_CreateVTable(CPpmd7z_RangeDec *p)
-{
-  p->vt.GetThreshold = Range_GetThreshold;
-  p->vt.Decode = Range_Decode;
-  p->vt.DecodeBit = Range_DecodeBit;
-}
 
 
-#define MASK(sym) ((signed char *)charMask)[sym]
+#define CTX(ref) ((CPpmd7_Context *)Ppmd7_GetContext(p, ref))
+typedef CPpmd7_Context * CTX_PTR;
+#define SUCCESSOR(p) Ppmd_GET_SUCCESSOR(p)
+void Ppmd7_UpdateModel(CPpmd7 *p);
 
-int Ppmd7_DecodeSymbol(CPpmd7 *p, const IPpmd7_RangeDec *rc)
+#define MASK(sym) ((unsigned char *)charMask)[sym]
+// MY_FORCE_INLINE
+// static
+int Ppmd7z_DecodeSymbol(CPpmd7 *p)
 {
   size_t charMask[256 / sizeof(size_t)];
+
   if (p->MinContext->NumStats != 1)
   {
     CPpmd_State *s = Ppmd7_GetStats(p, p->MinContext);
     unsigned i;
     UInt32 count, hiCnt;
-    if ((count = rc->GetThreshold(rc, p->MinContext->SummFreq)) < (hiCnt = s->Freq))
+    UInt32 summFreq = p->MinContext->Union2.SummFreq;
+
+    
+    
+    
+    count = RC_GetThreshold(summFreq);
+    hiCnt = count;
+    
+    if ((Int32)(count -= s->Freq) < 0)
     {
-      Byte symbol;
+      Byte sym;
-      rc->Decode(rc, 0, s->Freq);
+      RC_DecodeFinal(0, s->Freq);
       p->FoundState = s;
-      symbol = s->Symbol;
+      sym = s->Symbol;
       Ppmd7_Update1_0(p);
-      return symbol;
+      return sym;
     }
+  
     p->PrevSuccess = 0;
-    i = p->MinContext->NumStats - 1;
+    i = (unsigned)p->MinContext->NumStats - 1;
+    
     do
     {
-      if ((hiCnt += (++s)->Freq) > count)
+      if ((Int32)(count -= (++s)->Freq) < 0)
       {
-        Byte symbol;
+        Byte sym;
-        rc->Decode(rc, hiCnt - s->Freq, s->Freq);
+        RC_DecodeFinal((hiCnt - count) - s->Freq, s->Freq);
         p->FoundState = s;
-        symbol = s->Symbol;
+        sym = s->Symbol;
         Ppmd7_Update1(p);
-        return symbol;
+        return sym;
       }
     }
     while (--i);
-    if (count >= p->MinContext->SummFreq)
+    
-      return -2;
+    if (hiCnt >= summFreq)
-    p->HiBitsFlag = p->HB2Flag[p->FoundState->Symbol];
+      return PPMD7_SYM_ERROR;
-    rc->Decode(rc, hiCnt, p->MinContext->SummFreq - hiCnt);
+    
+    hiCnt -= count;
+    RC_Decode(hiCnt, summFreq - hiCnt);
+
+    p->HiBitsFlag = PPMD7_HiBitsFlag_3(p->FoundState->Symbol);
     PPMD_SetAllBitsIn256Bytes(charMask);
-    MASK(s->Symbol) = 0;
+    // i = p->MinContext->NumStats - 1;
-    i = p->MinContext->NumStats - 1;
+    // do { MASK((--s)->Symbol) = 0; } while (--i);
-    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
   {
+    CPpmd_State *s = Ppmd7Context_OneState(p->MinContext);
     UInt16 *prob = Ppmd7_GetBinSumm(p);
-    if (rc->DecodeBit(rc, *prob) == 0)
+    UInt32 pr = *prob;
+    UInt32 size0 = (R->Range >> 14) * pr;
+    pr = PPMD_UPDATE_PROB_1(pr);
+
+    if (R->Code < size0)
     {
-      Byte symbol;
+      Byte sym;
-      *prob = (UInt16)PPMD_UPDATE_PROB_0(*prob);
+      *prob = (UInt16)(pr + (1 << PPMD_INT_BITS));
-      symbol = (p->FoundState = Ppmd7Context_OneState(p->MinContext))->Symbol;
+      
-      Ppmd7_UpdateBin(p);
+      // RangeDec_DecodeBit0(size0);
-      return symbol;
+      R->Range = size0;
+      RC_NORM_1(R)
+      /* we can use single byte normalization here because of
+         (min(BinSumm[][]) = 95) > (1 << (14 - 8)) */
+
+      // sym = (p->FoundState = Ppmd7Context_OneState(p->MinContext))->Symbol;
+      // Ppmd7_UpdateBin(p);
+      {
+        unsigned freq = s->Freq;
+        CTX_PTR c = CTX(SUCCESSOR(s));
+        sym = s->Symbol;
+        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 sym;
     }
-    *prob = (UInt16)PPMD_UPDATE_PROB_1(*prob);
+
-    p->InitEsc = PPMD7_kExpEscape[*prob >> 10];
+    *prob = (UInt16)pr;
+    p->InitEsc = p->ExpEscape[pr >> 10];
+
+    // RangeDec_DecodeBit1(size0);
+    
+    R->Code -= size0;
+    R->Range -= size0;
+    RC_NORM_LOCAL(R)
+    
     PPMD_SetAllBitsIn256Bytes(charMask);
     MASK(Ppmd7Context_OneState(p->MinContext)->Symbol) = 0;
     p->PrevSuccess = 0;
   }
+
   for (;;)
   {
-    CPpmd_State *ps[256], *s;
+    CPpmd_State *s, *s2;
     UInt32 freqSum, count, hiCnt;
+
     CPpmd_See *see;
-    unsigned i, num, numMasked = p->MinContext->NumStats;
+    CPpmd7_Context *mc;
+    unsigned numMasked;
+    RC_NORM_REMOTE(R)
+    mc = p->MinContext;
+    numMasked = mc->NumStats;
+
     do
     {
       p->OrderFall++;
-      if (!p->MinContext->Suffix)
+      if (!mc->Suffix)
-        return -1;
+        return PPMD7_SYM_END;
-      p->MinContext = Ppmd7_GetContext(p, p->MinContext->Suffix);
+      mc = Ppmd7_GetContext(p, mc->Suffix);
     }
-    while (p->MinContext->NumStats == numMasked);
+    while (mc->NumStats == numMasked);
-    hiCnt = 0;
-    s = Ppmd7_GetStats(p, p->MinContext);
-    i = 0;
-    num = p->MinContext->NumStats - numMasked;
-    do
-    {
-      int k = (int)(MASK(s->Symbol));
-      hiCnt += (s->Freq & k);
-      ps[i] = s++;
-      i -= k;
-    }
-    while (i != num);
     
+    s = Ppmd7_GetStats(p, mc);
+
+    {
+      unsigned num = mc->NumStats;
+      unsigned num2 = num / 2;
+      
+      num &= 1;
+      hiCnt = (s->Freq & (unsigned)(MASK(s->Symbol))) & (0 - (UInt32)num);
+      s += num;
+      p->MinContext = mc;
+
+      do
+      {
+        unsigned sym0 = s[0].Symbol;
+        unsigned sym1 = s[1].Symbol;
+        s += 2;
+        hiCnt += (s[-2].Freq & (unsigned)(MASK(sym0)));
+        hiCnt += (s[-1].Freq & (unsigned)(MASK(sym1)));
+      }
+      while (--num2);
+    }
+
     see = Ppmd7_MakeEscFreq(p, numMasked, &freqSum);
     freqSum += hiCnt;
-    count = rc->GetThreshold(rc, freqSum);
+
+
+
+
+    count = RC_GetThreshold(freqSum);
     
     if (count < hiCnt)
     {
-      Byte symbol;
+      Byte sym;
-      CPpmd_State **pps = ps;
+
-      for (hiCnt = 0; (hiCnt += (*pps)->Freq) <= count; pps++);
+      s = Ppmd7_GetStats(p, p->MinContext);
-      s = *pps;
+      hiCnt = count;
-      rc->Decode(rc, hiCnt - s->Freq, s->Freq);
+      // count -= s->Freq & (unsigned)(MASK(s->Symbol));
+      // if ((Int32)count >= 0)
+      {
+        for (;;)
+        {
+          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);
+
+      // new (see->Summ) value can overflow over 16-bits in some rare cases
       Ppmd_See_Update(see);
       p->FoundState = s;
-      symbol = s->Symbol;
+      sym = s->Symbol;
       Ppmd7_Update2(p);
-      return symbol;
+      return sym;
     }
+
     if (count >= freqSum)
-      return -2;
+      return PPMD7_SYM_ERROR;
-    rc->Decode(rc, 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
     see->Summ = (UInt16)(see->Summ + freqSum);
-    do { MASK(ps[--i]->Symbol) = 0; } while (i != 0);
+
+    s = Ppmd7_GetStats(p, p->MinContext);
+    s2 = s + p->MinContext->NumStats;
+    do
+    {
+      MASK(s->Symbol) = 0;
+      s++;
+    }
+    while (s != s2);
   }
 }
+
+/*
+Byte *Ppmd7z_DecodeSymbols(CPpmd7 *p, Byte *buf, const Byte *lim)
+{
+  int sym = 0;
+  if (buf != lim)
+  do
+  {
+    sym = Ppmd7z_DecodeSymbol(p);
+    if (sym < 0)
+      break;
+    *buf = (Byte)sym;
+  }
+  while (++buf < lim);
+  p->LastSymbol = sym;
+  return buf;
+}
+*/

libraries/lzma/C/Threads.c

@@ -1,9 +1,11 @@
 /* Threads.c -- multithreading library
-2017-06-26 : Igor Pavlov : Public domain */
+2021-12-21 : Igor Pavlov : Public domain */
 
 #include "Precomp.h"
 
-#ifndef UNDER_CE
+#ifdef _WIN32
+
+#ifndef USE_THREADS_CreateThread
 #include <process.h>
 #endif
 
@@ -29,28 +31,103 @@ WRes HandlePtr_Close(HANDLE *p)
   return 0;
 }
 
-WRes Handle_WaitObject(HANDLE h) { return (WRes)WaitForSingleObject(h, INFINITE); }
+WRes Handle_WaitObject(HANDLE h)
+{
+  DWORD dw = WaitForSingleObject(h, INFINITE);
+  /*
+    (dw) result:
+    WAIT_OBJECT_0  // 0
+    WAIT_ABANDONED // 0x00000080 : is not compatible with Win32 Error space
+    WAIT_TIMEOUT   // 0x00000102 : is     compatible with Win32 Error space
+    WAIT_FAILED    // 0xFFFFFFFF
+  */
+  if (dw == WAIT_FAILED)
+  {
+    dw = GetLastError();
+    if (dw == 0)
+      return WAIT_FAILED;
+  }
+  return (WRes)dw;
+}
+
+#define Thread_Wait(p) Handle_WaitObject(*(p))
+
+WRes Thread_Wait_Close(CThread *p)
+{
+  WRes res = Thread_Wait(p);
+  WRes res2 = Thread_Close(p);
+  return (res != 0 ? res : res2);
+}
 
 WRes Thread_Create(CThread *p, THREAD_FUNC_TYPE func, LPVOID param)
 {
   /* Windows Me/98/95: threadId parameter may not be NULL in _beginthreadex/CreateThread functions */
-  
+
-  #ifdef UNDER_CE
+  #ifdef USE_THREADS_CreateThread
-  
+
   DWORD threadId;
-  *p = CreateThread(0, 0, func, param, 0, &threadId);
+  *p = CreateThread(NULL, 0, func, param, 0, &threadId);
-
+  
   #else
-
+  
   unsigned threadId;
-  *p = (HANDLE)_beginthreadex(NULL, 0, func, param, 0, &threadId);
+  *p = (HANDLE)(_beginthreadex(NULL, 0, func, param, 0, &threadId));
-   
+  
   #endif
 
   /* maybe we must use errno here, but probably GetLastError() is also OK. */
   return HandleToWRes(*p);
 }
 
+
+WRes Thread_Create_With_Affinity(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, CAffinityMask affinity)
+{
+  #ifdef USE_THREADS_CreateThread
+
+  UNUSED_VAR(affinity)
+  return Thread_Create(p, func, param);
+  
+  #else
+  
+  /* Windows Me/98/95: threadId parameter may not be NULL in _beginthreadex/CreateThread functions */
+  HANDLE h;
+  WRes wres;
+  unsigned threadId;
+  h = (HANDLE)(_beginthreadex(NULL, 0, func, param, CREATE_SUSPENDED, &threadId));
+  *p = h;
+  wres = HandleToWRes(h);
+  if (h)
+  {
+    {
+      // DWORD_PTR prevMask =
+      SetThreadAffinityMask(h, (DWORD_PTR)affinity);
+      /*
+      if (prevMask == 0)
+      {
+        // affinity change is non-critical error, so we can ignore it
+        // wres = GetError();
+      }
+      */
+    }
+    {
+      DWORD prevSuspendCount = ResumeThread(h);
+      /* ResumeThread() returns:
+         0 : was_not_suspended
+         1 : was_resumed
+        -1 : error
+      */
+      if (prevSuspendCount == (DWORD)-1)
+        wres = GetError();
+    }
+  }
+
+  /* maybe we must use errno here, but probably GetLastError() is also OK. */
+  return wres;
+
+  #endif
+}
+
+
 static WRes Event_Create(CEvent *p, BOOL manualReset, int signaled)
 {
   *p = CreateEvent(NULL, manualReset, (signaled ? TRUE : FALSE), NULL);
@@ -68,10 +145,22 @@ WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p) { return AutoResetEven
 
 WRes Semaphore_Create(CSemaphore *p, UInt32 initCount, UInt32 maxCount)
 {
+  // negative ((LONG)maxCount) is not supported in WIN32::CreateSemaphore()
   *p = CreateSemaphore(NULL, (LONG)initCount, (LONG)maxCount, NULL);
   return HandleToWRes(*p);
 }
 
+WRes Semaphore_OptCreateInit(CSemaphore *p, UInt32 initCount, UInt32 maxCount)
+{
+  // if (Semaphore_IsCreated(p))
+  {
+    WRes wres = Semaphore_Close(p);
+    if (wres != 0)
+      return wres;
+  }
+  return Semaphore_Create(p, initCount, maxCount);
+}
+
 static WRes Semaphore_Release(CSemaphore *p, LONG releaseCount, LONG *previousCount)
   { return BOOLToWRes(ReleaseSemaphore(*p, releaseCount, previousCount)); }
 WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 num)
@@ -80,7 +169,9 @@ WRes Semaphore_Release1(CSemaphore *p) { return Semaphore_ReleaseN(p, 1); }
 
 WRes CriticalSection_Init(CCriticalSection *p)
 {
-  /* InitializeCriticalSection can raise only STATUS_NO_MEMORY exception */
+  /* InitializeCriticalSection() can raise exception:
+     Windows XP, 2003 : can raise a STATUS_NO_MEMORY exception
+     Windows Vista+   : no exceptions */
   #ifdef _MSC_VER
   __try
   #endif
@@ -89,7 +180,361 @@ WRes CriticalSection_Init(CCriticalSection *p)
     /* InitializeCriticalSectionAndSpinCount(p, 0); */
   }
   #ifdef _MSC_VER
-  __except (EXCEPTION_EXECUTE_HANDLER) { return 1; }
+  __except (EXCEPTION_EXECUTE_HANDLER) { return ERROR_NOT_ENOUGH_MEMORY; }
   #endif
   return 0;
 }
+
+
+
+
+#else // _WIN32
+
+// ---------- POSIX ----------
+
+#ifndef __APPLE__
+#ifndef _7ZIP_AFFINITY_DISABLE
+// _GNU_SOURCE can be required for pthread_setaffinity_np() / CPU_ZERO / CPU_SET
+#define _GNU_SOURCE
+#endif
+#endif
+
+#include "Threads.h"
+
+#include <errno.h>
+#include <stdlib.h>
+#include <string.h>
+#ifdef _7ZIP_AFFINITY_SUPPORTED
+// #include <sched.h>
+#endif
+
+
+// #include <stdio.h>
+// #define PRF(p) p
+#define PRF(p)
+
+#define Print(s) PRF(printf("\n%s\n", s))
+
+// #include <stdio.h>
+
+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");
+
+  pthread_attr_t attr;
+  int ret;
+  // int ret2;
+
+  p->_created = 0;
+
+  RINOK(pthread_attr_init(&attr));
+
+  ret = pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_JOINABLE);
+
+  if (!ret)
+  {
+    if (cpuSet)
+    {
+      #ifdef _7ZIP_AFFINITY_SUPPORTED
+      
+      /*
+      printf("\n affinity :");
+      unsigned i;
+      for (i = 0; i < sizeof(*cpuSet) && i < 8; i++)
+      {
+        Byte b = *((const Byte *)cpuSet + i);
+        char temp[32];
+        #define GET_HEX_CHAR(t) ((char)(((t < 10) ? ('0' + t) : ('A' + (t - 10)))))
+        temp[0] = GET_HEX_CHAR((b & 0xF));
+        temp[1] = GET_HEX_CHAR((b >> 4));
+        // temp[0] = GET_HEX_CHAR((b >> 4));  // big-endian
+        // temp[1] = GET_HEX_CHAR((b & 0xF));  // big-endian
+        temp[2] = 0;
+        printf("%s", temp);
+      }
+      printf("\n");
+      */
+
+      // ret2 =
+      pthread_attr_setaffinity_np(&attr, sizeof(*cpuSet), cpuSet);
+      // if (ret2) ret = ret2;
+      #endif
+    }
+    
+    ret = pthread_create(&p->_tid, &attr, func, param);
+    
+    if (!ret)
+    {
+      p->_created = 1;
+      /*
+      if (cpuSet)
+      {
+        // ret2 =
+        pthread_setaffinity_np(p->_tid, sizeof(*cpuSet), cpuSet);
+        // if (ret2) ret = ret2;
+      }
+      */
+    }
+  }
+  // ret2 =
+  pthread_attr_destroy(&attr);
+  // if (ret2 != 0) ret = ret2;
+  return ret;
+}
+
+
+WRes Thread_Create(CThread *p, THREAD_FUNC_TYPE func, LPVOID param)
+{
+  return Thread_Create_With_CpuSet(p, func, param, NULL);
+}
+
+
+WRes Thread_Create_With_Affinity(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, CAffinityMask affinity)
+{
+  Print("Thread_Create_WithAffinity");
+  CCpuSet cs;
+  unsigned i;
+  CpuSet_Zero(&cs);
+  for (i = 0; i < sizeof(affinity) * 8; i++)
+  {
+    if (affinity == 0)
+      break;
+    if (affinity & 1)
+    {
+      CpuSet_Set(&cs, i);
+    }
+    affinity >>= 1;
+  }
+  return Thread_Create_With_CpuSet(p, func, param, &cs);
+}
+
+
+WRes Thread_Close(CThread *p)
+{
+  // Print("Thread_Close");
+  int ret;
+  if (!p->_created)
+    return 0;
+    
+  ret = pthread_detach(p->_tid);
+  p->_tid = 0;
+  p->_created = 0;
+  return ret;
+}
+
+
+WRes Thread_Wait_Close(CThread *p)
+{
+  // Print("Thread_Wait_Close");
+  void *thread_return;
+  int ret;
+  if (!p->_created)
+    return EINVAL;
+
+  ret = pthread_join(p->_tid, &thread_return);
+  // probably we can't use that (_tid) after pthread_join(), so we close thread here
+  p->_created = 0;
+  p->_tid = 0;
+  return ret;
+}
+
+
+
+static WRes Event_Create(CEvent *p, int manualReset, int signaled)
+{
+  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;
+  return 0;
+}
+
+WRes ManualResetEvent_Create(CManualResetEvent *p, int signaled)
+  { return Event_Create(p, True, signaled); }
+WRes ManualResetEvent_CreateNotSignaled(CManualResetEvent *p)
+  { return ManualResetEvent_Create(p, 0); }
+WRes AutoResetEvent_Create(CAutoResetEvent *p, int signaled)
+  { return Event_Create(p, False, signaled); }
+WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p)
+  { return AutoResetEvent_Create(p, 0); }
+
+
+WRes Event_Set(CEvent *p)
+{
+  RINOK(pthread_mutex_lock(&p->_mutex));
+  p->_state = True;
+  int res1 = pthread_cond_broadcast(&p->_cond);
+  int res2 = pthread_mutex_unlock(&p->_mutex);
+  return (res2 ? res2 : res1);
+}
+
+WRes Event_Reset(CEvent *p)
+{
+  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));
+  while (p->_state == False)
+  {
+    // ETIMEDOUT
+    // ret =
+    pthread_cond_wait(&p->_cond, &p->_mutex);
+    // if (ret != 0) break;
+  }
+  if (p->_manual_reset == False)
+  {
+    p->_state = False;
+  }
+  return pthread_mutex_unlock(&p->_mutex);
+}
+
+WRes Event_Close(CEvent *p)
+{
+  if (!p->_created)
+    return 0;
+  p->_created = 0;
+  {
+    int res1 = pthread_mutex_destroy(&p->_mutex);
+    int res2 = pthread_cond_destroy(&p->_cond);
+    return (res1 ? res1 : res2);
+  }
+}
+
+
+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));
+  p->_count = initCount;
+  p->_maxCount = maxCount;
+  p->_created = 1;
+  return 0;
+}
+
+
+WRes Semaphore_OptCreateInit(CSemaphore *p, UInt32 initCount, UInt32 maxCount)
+{
+  if (Semaphore_IsCreated(p))
+  {
+    /*
+    WRes wres = Semaphore_Close(p);
+    if (wres != 0)
+      return wres;
+    */
+    if (initCount > maxCount || maxCount < 1)
+      return EINVAL;
+    // return EINVAL; // for debug
+    p->_count = initCount;
+    p->_maxCount = maxCount;
+    return 0;
+  }
+  return Semaphore_Create(p, initCount, maxCount);
+}
+
+
+WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 releaseCount)
+{
+  UInt32 newCount;
+  int ret;
+
+  if (releaseCount < 1)
+    return EINVAL;
+
+  RINOK(pthread_mutex_lock(&p->_mutex));
+
+  newCount = p->_count + releaseCount;
+  if (newCount > p->_maxCount)
+    ret = ERROR_TOO_MANY_POSTS; // EINVAL;
+  else
+  {
+    p->_count = newCount;
+    ret = pthread_cond_broadcast(&p->_cond);
+  }
+  RINOK(pthread_mutex_unlock(&p->_mutex));
+  return ret;
+}
+
+WRes Semaphore_Wait(CSemaphore *p)
+{
+  RINOK(pthread_mutex_lock(&p->_mutex));
+  while (p->_count < 1)
+  {
+    pthread_cond_wait(&p->_cond, &p->_mutex);
+  }
+  p->_count--;
+  return pthread_mutex_unlock(&p->_mutex);
+}
+
+WRes Semaphore_Close(CSemaphore *p)
+{
+  if (!p->_created)
+    return 0;
+  p->_created = 0;
+  {
+    int res1 = pthread_mutex_destroy(&p->_mutex);
+    int res2 = pthread_cond_destroy(&p->_cond);
+    return (res1 ? res1 : res2);
+  }
+}
+
+
+
+WRes CriticalSection_Init(CCriticalSection *p)
+{
+  // Print("CriticalSection_Init");
+  if (!p)
+    return EINTR;
+  return pthread_mutex_init(&p->_mutex, NULL);
+}
+
+void CriticalSection_Enter(CCriticalSection *p)
+{
+  // Print("CriticalSection_Enter");
+  if (p)
+  {
+    // int ret =
+    pthread_mutex_lock(&p->_mutex);
+  }
+}
+
+void CriticalSection_Leave(CCriticalSection *p)
+{
+  // Print("CriticalSection_Leave");
+  if (p)
+  {
+    // int ret =
+    pthread_mutex_unlock(&p->_mutex);
+  }
+}
+
+void CriticalSection_Delete(CCriticalSection *p)
+{
+  // Print("CriticalSection_Delete");
+  if (p)
+  {
+    // int ret =
+    pthread_mutex_destroy(&p->_mutex);
+  }
+}
+
+LONG InterlockedIncrement(LONG volatile *addend)
+{
+  // Print("InterlockedIncrement");
+  #ifdef USE_HACK_UNSAFE_ATOMIC
+    LONG val = *addend + 1;
+    *addend = val;
+    return val;
+  #else
+    return __sync_add_and_fetch(addend, 1);
+  #endif
+}
+
+#endif // _WIN32

libraries/lzma/C/Threads.h

@@ -1,38 +1,139 @@
 /* Threads.h -- multithreading library
-2017-06-18 : Igor Pavlov : Public domain */
+2021-12-21 : Igor Pavlov : Public domain */
 
 #ifndef __7Z_THREADS_H
 #define __7Z_THREADS_H
 
 #ifdef _WIN32
-#include <windows.h>
+#include <Windows.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")
+// #define _GNU_SOURCE
+#endif
+#endif
+#endif
+
+#include <pthread.h>
+
 #endif
 
 #include "7zTypes.h"
 
 EXTERN_C_BEGIN
 
+#ifdef _WIN32
+
 WRes HandlePtr_Close(HANDLE *h);
 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))
+// #define Thread_Wait(p) Handle_WaitObject(*(p))
+
+#ifdef UNDER_CE
+  // if (USE_THREADS_CreateThread is      defined), we use _beginthreadex()
+  // if (USE_THREADS_CreateThread is not definned), we use CreateThread()
+  #define USE_THREADS_CreateThread
+#endif
 
 typedef
-#ifdef UNDER_CE
+    #ifdef USE_THREADS_CreateThread
-  DWORD
+      DWORD
+    #else
+      unsigned
+    #endif
+    THREAD_FUNC_RET_TYPE;
+
+typedef DWORD_PTR CAffinityMask;
+typedef DWORD_PTR CCpuSet;
+
+#define CpuSet_Zero(p) { *(p) = 0; }
+#define CpuSet_Set(p, cpu) { *(p) |= ((DWORD_PTR)1 << (cpu)); }
+
+#else //  _WIN32
+
+typedef struct _CThread
+{
+  pthread_t _tid;
+  int _created;
+} CThread;
+
+#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;
+
+typedef UInt64 CAffinityMask;
+
+#ifdef _7ZIP_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)
+
 #else
-  unsigned
+
+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)
+
 #endif
-  THREAD_FUNC_RET_TYPE;
+
+
+#endif //  _WIN32
+
 
 #define THREAD_FUNC_CALL_TYPE MY_STD_CALL
-#define THREAD_FUNC_DECL THREAD_FUNC_RET_TYPE THREAD_FUNC_CALL_TYPE
+
+#if defined(_WIN32) && defined(__GNUC__)
+/* GCC compiler for x86 32-bit uses the rule:
+   the stack is 16-byte aligned before CALL instruction for function calling.
+   But only root function main() contains instructions that
+   set 16-byte alignment for stack pointer. And another functions
+   just keep alignment, if it was set in some parent function.
+   
+   The problem:
+    if we create new thread in MinGW (GCC) 32-bit x86 via _beginthreadex() or CreateThread(),
+       the root function of thread doesn't set 16-byte alignment.
+       And stack frames in all child functions also will be unaligned in that case.
+   
+   Here we set (force_align_arg_pointer) attribute for root function of new thread.
+   Do we need (force_align_arg_pointer) also for another systems?  */
+  
+  #define THREAD_FUNC_ATTRIB_ALIGN_ARG __attribute__((force_align_arg_pointer))
+  // #define THREAD_FUNC_ATTRIB_ALIGN_ARG // for debug : bad alignment in SSE functions
+#else
+  #define THREAD_FUNC_ATTRIB_ALIGN_ARG
+#endif
+
+#define THREAD_FUNC_DECL  THREAD_FUNC_ATTRIB_ALIGN_ARG THREAD_FUNC_RET_TYPE THREAD_FUNC_CALL_TYPE
+
 typedef THREAD_FUNC_RET_TYPE (THREAD_FUNC_CALL_TYPE * THREAD_FUNC_TYPE)(void *);
 WRes Thread_Create(CThread *p, THREAD_FUNC_TYPE func, LPVOID param);
+WRes Thread_Create_With_Affinity(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, CAffinityMask affinity);
+WRes Thread_Wait_Close(CThread *p);
+
+#ifdef _WIN32
+#define Thread_Create_With_CpuSet(p, func, param, cs) \
+  Thread_Create_With_Affinity(p, func, param, *cs)
+#else
+WRes Thread_Create_With_CpuSet(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, const CCpuSet *cpuSet);
+#endif
+
+
+#ifdef _WIN32
 
 typedef HANDLE CEvent;
 typedef CEvent CAutoResetEvent;
@@ -54,6 +155,7 @@ typedef HANDLE CSemaphore;
 #define Semaphore_Close(p) HandlePtr_Close(p)
 #define Semaphore_Wait(p) Handle_WaitObject(*(p))
 WRes Semaphore_Create(CSemaphore *p, UInt32 initCount, UInt32 maxCount);
+WRes Semaphore_OptCreateInit(CSemaphore *p, UInt32 initCount, UInt32 maxCount);
 WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 num);
 WRes Semaphore_Release1(CSemaphore *p);
 
@@ -63,6 +165,68 @@ WRes CriticalSection_Init(CCriticalSection *p);
 #define CriticalSection_Enter(p) EnterCriticalSection(p)
 #define CriticalSection_Leave(p) LeaveCriticalSection(p)
 
+
+#else // _WIN32
+
+typedef struct _CEvent
+{
+  int _created;
+  int _manual_reset;
+  int _state;
+  pthread_mutex_t _mutex;
+  pthread_cond_t _cond;
+} CEvent;
+
+typedef CEvent CAutoResetEvent;
+typedef CEvent CManualResetEvent;
+
+#define Event_Construct(p) (p)->_created = 0
+#define Event_IsCreated(p) ((p)->_created)
+
+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);
+WRes Event_Close(CEvent *p);
+
+
+typedef struct _CSemaphore
+{
+  int _created;
+  UInt32 _count;
+  UInt32 _maxCount;
+  pthread_mutex_t _mutex;
+  pthread_cond_t _cond;
+} CSemaphore;
+
+#define Semaphore_Construct(p) (p)->_created = 0
+#define Semaphore_IsCreated(p) ((p)->_created)
+
+WRes Semaphore_Create(CSemaphore *p, UInt32 initCount, UInt32 maxCount);
+WRes Semaphore_OptCreateInit(CSemaphore *p, UInt32 initCount, UInt32 maxCount);
+WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 num);
+#define Semaphore_Release1(p) Semaphore_ReleaseN(p, 1)
+WRes Semaphore_Wait(CSemaphore *p);
+WRes Semaphore_Close(CSemaphore *p);
+
+
+typedef struct _CCriticalSection
+{
+  pthread_mutex_t _mutex;
+} CCriticalSection;
+
+WRes CriticalSection_Init(CCriticalSection *p);
+void CriticalSection_Delete(CCriticalSection *cs);
+void CriticalSection_Enter(CCriticalSection *cs);
+void CriticalSection_Leave(CCriticalSection *cs);
+
+LONG InterlockedIncrement(LONG volatile *addend);
+
+#endif  // _WIN32
+
 EXTERN_C_END
 
 #endif

libraries/lzma/CMakeLists.txt

@@ -4,7 +4,9 @@ make_release_only()
 
 set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -D_7ZIP_PPMD_SUPPPORT" )
 
-set( LZMA_FILES
+find_package(Threads)
+
+add_library( lzma STATIC
 	C/7zArcIn.c
 	C/7zBuf.c
 	C/7zCrc.c
@@ -18,17 +20,13 @@ set( LZMA_FILES
 	C/CpuArch.c
 	C/Delta.c
 	C/LzFind.c
+	C/LzFindMt.c
+	C/LzFindOpt.c
 	C/Lzma2Dec.c
 	C/LzmaDec.c
 	C/LzmaEnc.c
 	C/Ppmd7.c
-	C/Ppmd7Dec.c )
+	C/Ppmd7Dec.c
-
+	C/Threads.c
-if( WIN32 )
+)
-	set( LZMA_FILES ${LZMA_FILES} C/LzFindMt.c C/Threads.c )
+target_link_libraries( lzma Threads::Threads )
-else()
-	set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -D_7ZIP_ST" )
-endif()
-
-add_library( lzma STATIC ${LZMA_FILES} )
-target_link_libraries( lzma )

libraries/lzma/DOC/lzma-history.txt

@@ -1,6 +1,80 @@
 HISTORY of the LZMA SDK
 -----------------------
 
+21.07          2021-12-26
+-------------------------
+- New switches: -spm and -im!{file_path} to exclude directories from processing 
+    for specified paths that don't contain path separator character at the end of path.
+- The sorting order of files in archives was slightly changed to be more consistent
+  for cases where the name of some directory is the same as the prefix part of the name
+  of another directory or file.
+
+
+21.06          2021-11-24
+-------------------------
+- Bug in LZMA encoder in file LzmaEnc.c was fixed:
+  LzmaEnc_MemEncode(), LzmaEncode() and LzmaCompress() could work incorrectly, 
+    if size value for output buffer is smaller than size required for all compressed data.
+  LzmaEnc_Encode() could work incorrectly,
+    if callback ISeqOutStream::Write() doesn't write all compressed data.
+  NCompress::NLzma::CEncoder::Code() could work incorrectly,
+    if callback ISequentialOutStream::Write() returns error code.
+- Bug in versions 21.00-21.05 was fixed:
+  7-Zip didn't set attributes of directories during archive extracting.
+
+
+21.04 beta     2021-11-02
+-------------------------
+- 7-Zip now reduces the number of working CPU threads for compression,
+  if RAM size is not enough for compression with big LZMA2 dictionary.
+- 7-Zip now can create and check "file.sha256" text files that contain the list 
+  of file names and SHA-256 checksums in format compatible with sha256sum program.
+
+
+21.03 beta     2021-07-20
+-------------------------
+- The maximum dictionary size for LZMA/LZMA2 compressing was increased to 4 GB (3840 MiB).
+- Minor speed optimizations in LZMA/LZMA2 compressing.
+
+
+21.02 alpha    2021-05-06
+-------------------------
+- The command line version of 7-Zip for macOS was released.
+- The speed for LZMA and LZMA2 decompression in arm64 versions for macOS and Linux 
+  was increased by 20%-60%.
+
+
+21.01 alpha    2021-03-09
+-------------------------
+- The command line version of 7-Zip for Linux was released.
+- The improvements for speed of ARM64 version using hardware CPU instructions 
+  for AES, CRC-32, SHA-1 and SHA-256.
+- Some bugs were fixed.
+
+
+20.02 alpha    2020-08-08
+-------------------------
+- The default number of LZMA2 chunks per solid block in 7z archive was increased to 64.
+  It allows to increase the compression speed for big 7z archives, if there is a big number 
+  of CPU cores and threads.
+- The speed of PPMd compressing/decompressing was increased for 7z archives.
+- The new -ssp switch. If the switch -ssp is specified, 7-Zip doesn't allow the system 
+  to modify "Last Access Time" property of source files for archiving and hashing operations. 
+- Some bugs were fixed.
+
+
+20.00 alpha    2020-02-06
+-------------------------
+- 7-Zip now supports new optional match finders for LZMA/LZMA2 compression: bt5 and hc5, 
+  that can work faster than bt4 and hc4 match finders for the data with big redundancy.
+- The compression ratio was improved for Fast and Fastest compression levels with the 
+  following default settings:
+   - Fastest level (-mx1) : hc5 match finder with 256 KB dictionary.
+   - Fast    level (-mx3) : hc5 match finder with 4 MB dictionary.
+- Minor speed optimizations in multithreaded LZMA/LZMA2 compression for Normal/Maximum/Ultra 
+  compression levels.
+
+
 19.00          2019-02-21
 -------------------------
 - Encryption strength for 7z archives was increased:

libraries/lzma/DOC/lzma-sdk.txt

@@ -1,4 +1,4 @@
-LZMA SDK 19.00
+LZMA SDK 21.07
 --------------
 
 LZMA SDK provides the documentation, samples, header files,
@@ -62,14 +62,61 @@ LZMA SDK Contents
 
 UNIX/Linux version 
 ------------------
-To compile C++ version of file->file LZMA encoding, go to directory
+There are several otpions to compile 7-Zip with different compilers: gcc and clang.
-CPP/7zip/Bundles/LzmaCon
+Also 7-Zip code contains two versions for some critical parts of code: in C and in Assembeler.
-and call make to recompile it:
+So if you compile the version with Assembeler code, you will get faster 7-Zip binary.
-  make -f makefile.gcc clean all
+
+7-Zip's assembler code uses the following syntax for different platforms:
+
+1) x86 and x86-64 (AMD64): MASM syntax. 
+   There are 2 programs that supports MASM syntax in Linux.
+'    'Asmc Macro Assembler and JWasm. But JWasm now doesn't support some 
+      cpu instructions used in 7-Zip.
+   So you must install Asmc Macro Assembler in Linux, if you want to compile fastest version
+   of 7-Zip  x86 and x86-64:
+     https://github.com/nidud/asmc
+
+2) arm64: GNU assembler for ARM64 with preprocessor. 
+   That systax of that arm64 assembler code in 7-Zip is supported by GCC and CLANG for ARM64.
+
+There are different binaries that can be compiled from 7-Zip source.
+There are 2 main files in folder for compiling:
+  makefile        - that can be used for compiling Windows version of 7-Zip with nmake command
+  makefile.gcc    - that can be used for compiling Linux/macOS versions of 7-Zip with make command
+
+At first you must change the current folder to folder that contains `makefile.gcc`:
+
+  cd CPP/7zip/Bundles/Alone7z
+
+Then you can compile `makefile.gcc` with the command:
+
+  make -j -f makefile.gcc
+
+Also there are additional "*.mak" files in folder "CPP/7zip/" that can be used to compile 
+7-Zip binaries with optimized code and optimzing options.
+
+To compile with GCC without assembler:
+  cd CPP/7zip/Bundles/Alone7z
+  make -j -f ../../cmpl_gcc.mak
+
+To compile with CLANG without assembler:
+  make -j -f ../../cmpl_clang.mak
+
+To compile 7-Zip for x86-64 with asmc assembler:
+  make -j -f ../../cmpl_gcc_x64.mak
+
+To compile 7-Zip for arm64 with assembler:
+  make -j -f ../../cmpl_gcc_arm64.mak
+
+To compile 7-Zip for arm64 for macOS:
+  make -j -f ../../cmpl_mac_arm64.mak
+
+Also you can change some compiler options in the mak files:
+  cmpl_gcc.mak
+  var_gcc.mak
+  warn_gcc.mak
+
 
-In some UNIX/Linux versions you must compile LZMA with static libraries.
-To compile with static libraries, you can use 
-LIB = -lm -static
 
 Also you can use p7zip (port of 7-Zip for POSIX systems like Unix or Linux):
   

src/CMakeLists.txt

@@ -937,7 +937,6 @@ set (PCH_SOURCES
 	rendering/hwrenderer/hw_vertexbuilder.cpp
 	rendering/hwrenderer/doom_aabbtree.cpp
 	rendering/hwrenderer/hw_models.cpp
-	rendering/hwrenderer/hw_postprocessshader.cpp
 	rendering/hwrenderer/hw_precache.cpp
 	rendering/hwrenderer/scene/hw_lighting.cpp
 	rendering/hwrenderer/scene/hw_drawlistadd.cpp
@@ -1152,8 +1151,10 @@ set (PCH_SOURCES
 	common/rendering/hwrenderer/data/hw_aabbtree.cpp
 	common/rendering/hwrenderer/data/hw_shadowmap.cpp
 	common/rendering/hwrenderer/data/hw_shaderpatcher.cpp
+	common/rendering/hwrenderer/postprocessing/hw_postprocessshader.cpp
 	common/rendering/hwrenderer/postprocessing/hw_postprocess.cpp
 	common/rendering/hwrenderer/postprocessing/hw_postprocess_cvars.cpp
+	common/rendering/hwrenderer/postprocessing/hw_postprocessshader_ccmds.cpp
 	common/rendering/gl_load/gl_interface.cpp
 	common/rendering/gl/gl_renderer.cpp
 	common/rendering/gl/gl_stereo3d.cpp

src/common/2d/v_2ddrawer.cpp

@@ -749,7 +749,6 @@ void F2DDrawer::AddPoly(FGameTexture *texture, FVector2 *points, int npoints,
 void F2DDrawer::AddPoly(FGameTexture* img, FVector4* vt, size_t vtcount, const unsigned int* ind, size_t idxcount, int translation, PalEntry color, FRenderStyle style, int clipx1, int clipy1, int clipx2, int clipy2)
 {
 	RenderCommand dg;
-	int method = 0;
 
 	if (!img || !img->isValid()) return;
 
@@ -835,13 +834,13 @@ void F2DDrawer::AddFlatFill(int left, int top, int right, int bottom, FGameTextu
 	dg.mFlags = DTF_Wrap;
 
 	float fs = 1.f / float(flatscale);
-	bool flipc = false;
 
 	float sw = GetClassicFlatScalarWidth();
 	float sh = GetClassicFlatScalarHeight();
 
 	switch (local_origin)
 	{
+	default:
 	case 0:
 		fU1 = float(left) / (float)src->GetDisplayWidth() * fs;
 		fV1 = float(top) / (float)src->GetDisplayHeight() * fs;
@@ -994,7 +993,7 @@ void F2DDrawer::AddLine(double x1, double y1, double x2, double y2, int clipx1,
 		dg.mScissor[3] = clipy2 + 1 + int(offset.Y);
 		dg.mFlags |= DTF_Scissor;
 	}
-	
+
 	dg.mType = DrawTypeLines;
 	dg.mRenderStyle = LegacyRenderStyles[STYLE_Translucent];
 	dg.mVertCount = 2;

src/common/2d/v_2ddrawer.h

@@ -101,7 +101,7 @@ public:
 		}
 
 	};
-	
+
 	struct RenderCommand
 	{
 		EDrawType mType;
@@ -175,7 +175,7 @@ public:
 public:
 	int fullscreenautoaspect = 3;
 	int cliptop = -1, clipleft = -1, clipwidth = -1, clipheight = -1;
-	
+
 	int AddCommand(RenderCommand *data);
 	void AddIndices(int firstvert, int count, ...);
 private:
@@ -200,8 +200,8 @@ public:
 	void ClearScreen(PalEntry color = 0xff000000);
 	void AddDim(PalEntry color, float damount, int x1, int y1, int w, int h);
 	void AddClear(int left, int top, int right, int bottom, int palcolor, uint32_t color);
-	
+
-		
+
 	void AddLine(double x1, double y1, double x2, double y2, int cx, int cy, int cx2, int cy2, uint32_t color, uint8_t alpha = 255);
 	void AddThickLine(int x1, int y1, int x2, int y2, double thickness, uint32_t color, uint8_t alpha = 255);
 	void AddPixel(int x1, int y1, uint32_t color);

src/common/2d/v_draw.cpp

@@ -144,7 +144,7 @@ int GetUIScale(F2DDrawer *drawer, int altval)
 		// Default should try to scale to 640x400
 		int vscale = drawer->GetHeight() / 400;
 		int hscale = drawer->GetWidth() / 640;
-		scaleval = clamp(vscale, 1, hscale);
+		scaleval = max(1, min(vscale, hscale));
 	}
 	else scaleval = uiscale;
 
@@ -165,7 +165,7 @@ int GetConScale(F2DDrawer* drawer, int altval)
 		// Default should try to scale to 640x400
 		int vscale = drawer->GetHeight() / 800;
 		int hscale = drawer->GetWidth() / 1280;
-		scaleval = clamp(vscale, 1, hscale);
+		scaleval = max(1, min(vscale, hscale));
 	}
 	else scaleval = (uiscale+1) / 2;
 
@@ -671,7 +671,6 @@ bool ParseDrawTextureTags(F2DDrawer *drawer, FGameTexture *img, double x, double
 {
 	INTBOOL boolval;
 	int intval;
-	bool translationset = false;
 	bool fillcolorset = false;
 
 	if (!fortext)
@@ -858,7 +857,7 @@ bool ParseDrawTextureTags(F2DDrawer *drawer, FGameTexture *img, double x, double
 			parms->cleanmode = DTA_Base;
 			parms->virtHeight = ListGetDouble(tags);
 			break;
-			
+
 		case DTA_FullscreenScale:
 			intval = ListGetInt(tags);
 			if (intval >= FSMode_None && intval < FSMode_Max)
@@ -1016,10 +1015,16 @@ bool ParseDrawTextureTags(F2DDrawer *drawer, FGameTexture *img, double x, double
 		case DTA_CenterOffsetRel:
 			assert(fortext == false);
 			if (fortext) return false;
-			if (ListGetInt(tags))
+			intval = ListGetInt(tags);
+			if (intval == 1)
 			{
-				parms->left = img->GetDisplayLeftOffset() + img->GetDisplayWidth() * 0.5;
+				parms->left = img->GetDisplayLeftOffset() + (img->GetDisplayWidth() * 0.5);
-				parms->top = img->GetDisplayTopOffset() + img->GetDisplayHeight() * 0.5;
+				parms->top = img->GetDisplayTopOffset() + (img->GetDisplayHeight() * 0.5);
+			}
+			else if (intval == 2)
+			{
+				parms->left = img->GetDisplayLeftOffset() + floor(img->GetDisplayWidth() * 0.5);
+				parms->top = img->GetDisplayTopOffset() + floor(img->GetDisplayHeight() * 0.5);
 			}
 			break;
 

src/common/2d/v_draw.h

@@ -36,7 +36,7 @@ enum
 
 
 	FSMode_Max,
-	
+
 	// These all use ScaleToFit43, their purpose is to cut down on verbosity because they imply the virtual screen size.
 	FSMode_Predefined = 1000,
 	FSMode_Fit320x200 = 1000,
@@ -325,3 +325,5 @@ public:
 		CleanHeight = savedheight;
 	}
 };
+
+void Draw2D(F2DDrawer* drawer, FRenderState& state);

src/common/audio/music/i_music.cpp

@@ -226,7 +226,7 @@ void I_InitMusic(void)
 	nomusic = !!Args->CheckParm("-nomusic") || !!Args->CheckParm("-nosound");
 
 	snd_mididevice.Callback();
-	
+
 	ZMusicCallbacks callbacks{};
 
 	callbacks.MessageFunc = zmusic_printfunc;

src/common/audio/music/i_soundfont.cpp

@@ -140,7 +140,7 @@ FileReader FSoundFontReader::Open(const char *name, std::string& filename)
 ZMusicCustomReader* FSoundFontReader::open_interface(const char* name)
 {
 	std::string filename;
-	
+
 	FileReader fr = Open(name, filename);
 	if (!fr.isOpen()) return nullptr;
 	auto fri = GetMusicReader(fr);
@@ -336,7 +336,7 @@ void FSoundFontManager::ProcessOneFile(const FString &fn)
 		// We already got a soundfont with this name. Do not add again.
 		if (!sfi.mName.CompareNoCase(fb)) return;
 	}
-	
+
 	FileReader fr;
 	if (fr.OpenFile(fn))
 	{

src/common/audio/music/i_soundfont.h

@@ -29,13 +29,13 @@ protected:
     // When reading from an archive it will always be case insensitive, just like the lump manager.
     bool mCaseSensitivePaths = false;
     TArray<FString> mPaths;
-    
+
-  
+
     int pathcmp(const char *p1, const char *p2);
-   
+
-   
+
 public:
-    
+
     virtual ~FSoundFontReader() {}
     virtual FileReader OpenMainConfigFile() = 0;    // this is special because it needs to be synthesized for .sf files and set some restrictions for patch sets
 	virtual FString MainConfigFileName()
@@ -147,15 +147,15 @@ public:
 class FSoundFontManager
 {
     TArray<FSoundFontInfo> soundfonts;
-    
+
     void ProcessOneFile(const FString & fn);
-    
+
 public:
     void CollectSoundfonts();
     const FSoundFontInfo *FindSoundFont(const char *name, int allowedtypes) const;
     FSoundFontReader *OpenSoundFont(const char *name, int allowedtypes);
     const auto &GetList() const { return soundfonts; } // This is for the menu
-    
+
 };
 
 

src/common/audio/music/music.cpp

@@ -180,7 +180,7 @@ static bool FillStream(SoundStream* stream, void* buff, int len, void* userdata)
 			fbuf[i] = convert[i] * mus_playing.replayGainFactor * (1.f/32768.f);
 		}
 	}
-	
+
 	if (!written)
 	{
 		memset((char*)buff, 0, len);
@@ -298,7 +298,7 @@ void S_UpdateMusic ()
 	if (mus_playing.handle != nullptr)
 	{
 		ZMusic_Update(mus_playing.handle);
-		
+
 		// [RH] Update music and/or playlist. IsPlaying() must be called
 		// to attempt to reconnect to broken net streams and to advance the
 		// playlist when the current song finishes.
@@ -603,13 +603,13 @@ static void CheckReplayGain(const char *musicname, EMidiDevice playertype, const
 		{
 			float* sbuf = (float*)readbuffer.Data();
 			int numsamples = fmt.mBufferSize / 8;
-			auto index = lbuffer.Reserve(numsamples);
+			auto addr = lbuffer.Reserve(numsamples);
 			rbuffer.Reserve(numsamples);
 
 			for (int i = 0; i < numsamples; i++)
 			{
-				lbuffer[index + i] = sbuf[i * 2] * 32768.f;
+				lbuffer[addr + i] = sbuf[i * 2] * 32768.f;
-				rbuffer[index + i] = sbuf[i * 2 + 1] * 32768.f;
+				rbuffer[addr + i] = sbuf[i * 2 + 1] * 32768.f;
 			}
 		}
 		float accTime = lbuffer.Size() / (float)fmt.mSampleRate;
@@ -684,8 +684,6 @@ bool S_ChangeMusic(const char* musicname, int order, bool looping, bool force)
 		return true;
 	}
 
-	int lumpnum = -1;
-	int length = 0;
 	ZMusic_MusicStream handle = nullptr;
 	MidiDeviceSetting* devp = MidiDevices.CheckKey(musicname);
 

src/common/audio/sound/i_sound.h

@@ -107,7 +107,7 @@ public:
 
 	// Streaming sounds.
 	virtual SoundStream *CreateStream (SoundStreamCallback callback, int buffbytes, int flags, int samplerate, void *userdata) = 0;
-  
+
 	// Starts a sound.
 	virtual FISoundChannel *StartSound (SoundHandle sfx, float vol, int pitch, int chanflags, FISoundChannel *reuse_chan, float startTime = 0.f) = 0;
 	virtual FISoundChannel *StartSound3D (SoundHandle sfx, SoundListener *listener, float vol, FRolloffInfo *rolloff, float distscale, int pitch, int priority, const FVector3 &pos, const FVector3 &vel, int channum, int chanflags, FISoundChannel *reuse_chan, float startTime = 0.f) = 0;

src/common/audio/sound/oalsound.cpp

@@ -329,7 +329,6 @@ public:
 	virtual FString GetStats()
 		{
 		FString stats;
-		size_t pos = 0, len = 0;
 		ALfloat volume;
 		ALint offset;
 		ALint processed;
@@ -1676,7 +1675,7 @@ void OpenALSoundRenderer::UpdateListener(SoundListener *listener)
 
 		const_cast<ReverbContainer*>(env)->Modified = false;
 	}
-	
+
 	// NOTE: Moving into and out of water will undo pitch variations on sounds.
 	if(listener->underwater || env->SoftwareWater)
 	{
@@ -1722,7 +1721,7 @@ void OpenALSoundRenderer::UpdateListener(SoundListener *listener)
 	}
 	else if(WasInWater)
 	{
-		
+
 		WasInWater = false;
 
 		if(EnvSlot != 0)

src/common/audio/sound/s_environment.cpp

@@ -530,7 +530,7 @@ void S_ReadReverbDef (FScanner &sc)
 {
 	const ReverbContainer *def;
 	ReverbContainer *newenv;
-	REVERB_PROPERTIES props;
+	REVERB_PROPERTIES props = {};
 	char *name;
 	int id1, id2, i, j;
 	bool inited[NUM_REVERB_FIELDS];

src/common/audio/sound/s_reverbedit.cpp

@@ -309,7 +309,6 @@ DEFINE_ACTION_FUNCTION(DReverbEdit, GetValue)
 		}
 	}
 	ACTION_RETURN_FLOAT(v);
-	return 1;
 }
 
 DEFINE_ACTION_FUNCTION(DReverbEdit, SetValue)
@@ -337,14 +336,12 @@ DEFINE_ACTION_FUNCTION(DReverbEdit, SetValue)
 	}
 
 	ACTION_RETURN_FLOAT(v);
-	return 1;
 }
 
 DEFINE_ACTION_FUNCTION(DReverbEdit, GrayCheck)
 {
 	PARAM_PROLOGUE;
 	ACTION_RETURN_BOOL(CurrentEnv->Builtin);
-	return 1;
 }
 
 DEFINE_ACTION_FUNCTION(DReverbEdit, GetSelectedEnvironment)

src/common/audio/sound/s_sound.cpp

@@ -164,7 +164,6 @@ void SoundEngine::CacheSound (sfxinfo_t *sfx)
 {
 	if (GSnd && !sfx->bTentative)
 	{
-		sfxinfo_t *orig = sfx;
 		while (!sfx->bRandomHeader && sfx->link != sfxinfo_t::NO_LINK)
 		{
 			sfx = &S_sfx[sfx->link];
@@ -389,7 +388,7 @@ FSoundChan *SoundEngine::StartSound(int type, const void *source,
 	FVector3 pos, vel;
 	FRolloffInfo *rolloff;
 
-	if (sound_id <= 0 || volume <= 0 || nosfx || !SoundEnabled() || blockNewSounds)
+	if (sound_id <= 0 || volume <= 0 || nosfx || !SoundEnabled() || blockNewSounds || (unsigned)sound_id >= S_sfx.Size())
 		return NULL;
 
 	// prevent crashes.
@@ -403,7 +402,7 @@ FSoundChan *SoundEngine::StartSound(int type, const void *source,
 	{
 		return nullptr;
 	}
-	
+
 	sfx = &S_sfx[sound_id];
 
 	// Scale volume according to SNDINFO data.
@@ -612,7 +611,7 @@ FSoundChan *SoundEngine::StartSound(int type, const void *source,
 		{
 			chan->Source = source;
 		}
-		
+
 		if (spitch > 0.0)				// A_StartSound has top priority over all others.
 			SetPitch(chan, spitch);
 		else if (defpitch > 0.0)	// $PitchSet overrides $PitchShift
@@ -727,7 +726,7 @@ sfxinfo_t *SoundEngine::LoadSound(sfxinfo_t *sfx)
 		{
 			return sfx;
 		}
-		
+
 		// See if there is another sound already initialized with this lump. If so,
 		// then set this one up as a link, and don't load the sound again.
 		for (i = 0; i < S_sfx.Size(); i++)
@@ -831,7 +830,7 @@ bool SoundEngine::CheckSoundLimit(sfxinfo_t *sfx, const FVector3 &pos, int near_
 {
 	FSoundChan *chan;
 	int count;
-	
+
 	for (chan = Channels, count = 0; chan != NULL && count < near_limit; chan = chan->NextChan)
 	{
 		if (chan->ChanFlags & CHANF_FORGETTABLE) continue;
@@ -1085,13 +1084,14 @@ void SoundEngine::SetPitch(FSoundChan *chan, float pitch)
 // Is a sound being played by a specific emitter?
 //==========================================================================
 
-int SoundEngine::GetSoundPlayingInfo (int sourcetype, const void *source, int sound_id)
+int SoundEngine::GetSoundPlayingInfo (int sourcetype, const void *source, int sound_id, int chann)
 {
 	int count = 0;
 	if (sound_id > 0)
 	{
 		for (FSoundChan *chan = Channels; chan != NULL; chan = chan->NextChan)
 		{
+			if (chann != -1 && chann != chan->EntChannel) continue;
 			if (chan->OrgID == sound_id && (sourcetype == SOURCE_Any ||
 				(chan->SourceType == sourcetype &&
 				chan->Source == source)))
@@ -1104,6 +1104,7 @@ int SoundEngine::GetSoundPlayingInfo (int sourcetype, const void *source, int so
 	{
 		for (FSoundChan* chan = Channels; chan != NULL; chan = chan->NextChan)
 		{
+			if (chann != -1 && chann != chan->EntChannel) continue;
 			if ((sourcetype == SOURCE_Any || (chan->SourceType == sourcetype &&	chan->Source == source)))
 			{
 				count++;

src/common/audio/sound/s_soundinternal.h

@@ -73,7 +73,7 @@ class FSoundID
 {
 public:
 	FSoundID() = default;
-	
+
 	static FSoundID byResId(int ndx)
 	{
 		return FSoundID(S_FindSoundByResID(ndx)); 
@@ -120,7 +120,7 @@ protected:
 	enum EDummy { NoInit };
 	FSoundID(EDummy) {}
 };
- 
+
  class FSoundIDNoInit : public FSoundID
 {
 public:
@@ -196,7 +196,7 @@ void S_SetEnvironment (const ReverbContainer *settings);
 ReverbContainer *S_FindEnvironment (const char *name);
 ReverbContainer *S_FindEnvironment (int id);
 void S_AddEnvironment (ReverbContainer *settings);
-	
+
 class SoundEngine
 {
 protected:
@@ -249,9 +249,6 @@ public:
 		blockNewSounds = on;
 	}
 
-	virtual int SoundSourceIndex(FSoundChan* chan) { return 0; }
-	virtual void SetSource(FSoundChan* chan, int index) {}
-
 	virtual void StopChannel(FSoundChan* chan);
 	sfxinfo_t* LoadSound(sfxinfo_t* sfx);
 	const sfxinfo_t* GetSfx(unsigned snd)
@@ -303,7 +300,7 @@ public:
 	bool IsSourcePlayingSomething(int sourcetype, const void* actor, int channel, int sound_id = -1);
 
 	// Stop and resume music, during game PAUSE.
-	int GetSoundPlayingInfo(int sourcetype, const void* source, int sound_id);
+	int GetSoundPlayingInfo(int sourcetype, const void* source, int sound_id, int chan = -1);
 	void UnloadAllSounds();
 	void Reset();
 	void MarkUsed(int num);

src/common/console/c_bind.cpp

@@ -393,7 +393,7 @@ void FKeyBindings::PerformBind(FCommandLine &argv, const char *msg)
 	else
 	{
 		Printf ("%s:\n", msg);
-		
+
 		for (i = 0; i < NUM_KEYS; i++)
 		{
 			if (!Binds[i].IsEmpty())
@@ -696,7 +696,7 @@ void ReadBindings(int lump, bool override)
 			}
 			continue;
 		}
-		
+
 		// bind destination is optional and is the same as the console command
 		if (sc.Compare("bind"))
 		{

src/common/console/c_buttons.cpp

@@ -101,13 +101,13 @@ int ButtonMap::ListActionCommands (const char *pattern)
 int ButtonMap::FindButtonIndex (const char *key, int funclen) const
 {
     if (!key) return -1;
-	
+
 	FName name = funclen == -1? FName(key, true) : FName(key, funclen, true);
 	if (name == NAME_None) return -1;
-	
+
 	auto res = NameToNum.CheckKey(name);
 	if (!res) return -1;
-	
+
 	return *res;
 }
 

src/common/console/c_commandbuffer.cpp

@@ -299,7 +299,7 @@ void FCommandBuffer::AddString(FString clip)
 		}
 		auto strp = (const uint8_t*)clip.GetChars();
 		while (auto chr = GetCharFromString(strp)) build += chr;
-		
+
 		if (Text.length() == 0)
 		{
 			Text = build;

src/common/console/c_console.cpp

@@ -331,16 +331,16 @@ void C_DeinitConsole ()
 	// at runtime.)
 	for (size_t i = 0; i < countof(Commands); ++i)
 	{
-		FConsoleCommand *cmd = Commands[i];
+		FConsoleCommand *command = Commands[i];
 
-		while (cmd != NULL)
+		while (command != NULL)
 		{
-			FConsoleCommand *next = cmd->m_Next;
+			FConsoleCommand *nextcmd = command->m_Next;
-			if (cmd->IsAlias())
+			if (command->IsAlias())
 			{
-				delete cmd;
+				delete command;
 			}
-			cmd = next;
+			command = nextcmd;
 		}
 	}
 
@@ -1027,7 +1027,7 @@ static bool C_HandleKey (event_t *ev, FCommandBuffer &buffer)
 			TabbedList = false;
 			break;
 		}
-		
+
 		case '`':
 			// Check to see if we have ` bound to the console before accepting
 			// it as a way to close the console.

src/common/console/c_consolebuffer.cpp

@@ -72,7 +72,7 @@ FConsoleBuffer::FConsoleBuffer()
 void FConsoleBuffer::AddText(int printlevel, const char *text)
 {
 	FString build = TEXTCOLOR_TAN;
-	
+
 	if (mAddType == REPLACELINE)
 	{
 		// Just wondering: Do we actually need this case? If so, it may need some work.
@@ -85,15 +85,15 @@ void FConsoleBuffer::AddText(int printlevel, const char *text)
 		printlevel = -1;
 		mLastLineNeedsUpdate = true;
 	}
-	
+
 	if (printlevel >= 0 && printlevel != PRINT_HIGH)
 	{
 		if (printlevel == 200) build = TEXTCOLOR_GREEN;
 		else if (printlevel < PRINTLEVELS) build.Format("%c%c", TEXTCOLOR_ESCAPE, PrintColors[printlevel]+'A');
 	}
-	
+
 	size_t textsize = strlen(text);
-	
+
 	if (text[textsize-1] == '\r')
 	{
 		textsize--;

src/common/console/c_consolebuffer.h

@@ -56,12 +56,12 @@ class FConsoleBuffer
 	EAddType mAddType;
 	int mTextLines;
 	bool mBufferWasCleared;
-	
+
 	FFont *mLastFont;
 	int mLastDisplayWidth;
 	bool mLastLineNeedsUpdate;
 
-	
+
 public:
 	FConsoleBuffer();
 	void AddText(int printlevel, const char *string);

src/common/console/c_cvars.cpp

@@ -333,6 +333,7 @@ UCVarValue FBaseCVar::FromBool (bool value, ECVarType type)
 		break;
 
 	default:
+		ret.Int = 0;
 		break;
 	}
 
@@ -363,6 +364,7 @@ UCVarValue FBaseCVar::FromInt (int value, ECVarType type)
 		break;
 
 	default:
+		ret.Int = 0;
 		break;
 	}
 
@@ -395,6 +397,7 @@ UCVarValue FBaseCVar::FromFloat (float value, ECVarType type)
 		break;
 
 	default:
+		ret.Int = 0;
 		break;
 	}
 
@@ -456,6 +459,7 @@ UCVarValue FBaseCVar::FromString (const char *value, ECVarType type)
 		break;
 
 	default:
+		ret.Int = 0;
 		break;
 	}
 
@@ -1426,12 +1430,12 @@ void C_ArchiveCVars (FConfigFile *f, uint32_t filter)
 		cvar = cvar->m_Next;
 	}
 	qsort(cvarlist.Data(), cvarlist.Size(), sizeof(FBaseCVar*), cvarcmp);
-	for (auto cvar : cvarlist)
+	for (auto cv : cvarlist)
 	{
-		const char* const value = (cvar->Flags & CVAR_ISDEFAULT)
+		const char* const value = (cv->Flags & CVAR_ISDEFAULT)
-			? cvar->GetGenericRep(CVAR_String).String
+			? cv->GetGenericRep(CVAR_String).String
-			: cvar->SafeValue.GetChars();
+			: cv->SafeValue.GetChars();
-		f->SetValueForKey(cvar->GetName(), value);
+		f->SetValueForKey(cv->GetName(), value);
 	}
 }
 
@@ -1643,7 +1647,6 @@ CCMD (archivecvar)
 void C_ListCVarsWithoutDescription()
 {
 	FBaseCVar* var = CVars;
-	int count = 0;
 
 	while (var)
 	{

src/common/console/c_cvars.h

@@ -175,7 +175,7 @@ public:
 	static void ResetColors ();		// recalc color cvars' indices after screen change
 
 	static void ListVars (const char *filter, bool plain);
-	
+
 	const FString &GetDescription() const { return Description; };
 	const FString& GetToggleMessage(int which) { return ToggleMessages[which]; }
 	void SetToggleMessages(const char* on, const char* off)
@@ -398,7 +398,7 @@ public:
 
 protected:
 	virtual void DoSet (UCVarValue value, ECVarType type);
-	
+
 	static UCVarValue FromInt2 (int value, ECVarType type);
 	static int ToInt2 (UCVarValue value, ECVarType type);
 };

src/common/console/c_dispatch.cpp

@@ -253,7 +253,7 @@ void C_DoCommand (const char *cmd, int keynum)
 			return;
 		}
 	}
-	
+
 	// Parse it as a normal command
 	// Checking for matching commands follows this search order:
 	//	1. Check the Commands[] hash table
@@ -282,8 +282,8 @@ void C_DoCommand (const char *cmd, int keynum)
 			}
 			else
 			{
-				auto cmd = new FStoredCommand(com, beg);
+				auto command = new FStoredCommand(com, beg);
-				delayedCommandQueue.AddCommand(cmd);
+				delayedCommandQueue.AddCommand(command);
 			}
 		}
 	}
@@ -373,8 +373,8 @@ void AddCommandString (const char *text, int keynum)
 						  // Note that deferred commands lose track of which key
 						  // (if any) they were pressed from.
 							*brkpt = ';';
-							auto cmd = new FWaitingCommand(brkpt, tics, UnsafeExecutionContext);
+							auto command = new FWaitingCommand(brkpt, tics, UnsafeExecutionContext);
-							delayedCommandQueue.AddCommand(cmd);
+							delayedCommandQueue.AddCommand(command);
 						}
 						return;
 					}
@@ -851,8 +851,8 @@ CCMD (key)
 
 		for (i = 1; i < argv.argc(); ++i)
 		{
-			unsigned int key = MakeKey (argv[i]);
+			unsigned int hash = MakeKey (argv[i]);
-			Printf (" 0x%08x\n", key);
+			Printf (" 0x%08x\n", hash);
 		}
 	}
 }
@@ -1014,7 +1014,6 @@ void FExecList::AddPullins(TArray<FString> &wads, FConfigFile *config) const
 FExecList *C_ParseExecFile(const char *file, FExecList *exec)
 {
 	char cmd[4096];
-	int retval = 0;
 
 	FileReader fr;
 

src/common/console/c_expr.cpp

@@ -248,7 +248,7 @@ done:
 bool IsFloat (const char *str)
 {
 	const char *pt;
-	
+
 	if (*str == '+' || *str == '-')
 		str++;
 
@@ -367,7 +367,7 @@ static FStringProd *DoubleToString (FProduction *prod)
 static FDoubleProd *StringToDouble (FProduction *prod)
 {
 	FDoubleProd *newprod;
-	
+
 	newprod = NewDoubleProd (atof (static_cast<FStringProd *>(prod)->Value));
 	M_Free (prod);
 	return newprod;

src/common/console/c_notifybufferbase.cpp

@@ -126,7 +126,7 @@ void FNotifyBufferBase::Tick()
 	{
 		Text[i].Ticker++;
 	}
-	
+
 	for (i = 0; i < Text.Size(); ++i)
 	{
 		if (Text[i].TimeOut != 0 && Text[i].TimeOut > Text[i].Ticker)

src/common/engine/cycler.cpp

@@ -97,20 +97,20 @@ void FCycler::Update(double diff)
 {
 	double mult, angle;
 	double step = m_end - m_start;
-	
+
 	if (!m_shouldCycle)
 	{
 		return;
 	}
-	
+
 	m_time += diff;
 	if (m_time >= m_cycle)
 	{
 		m_time = m_cycle;
 	}
-	
+
 	mult = m_time / m_cycle;
-	
+
 	switch (m_cycleType)
 	{
 	case CYCLE_Linear:
@@ -149,7 +149,7 @@ void FCycler::Update(double diff)
 		}
 		break;
 	}
-	
+
 	if (m_time == m_cycle)
 	{
 		m_time = 0.;

src/common/engine/cycler.h

@@ -23,7 +23,7 @@ public:
 	FCycler() = default;
 	FCycler(const FCycler &other) = default;
 	FCycler &operator=(const FCycler &other) = default;
-	
+
    void Update(double diff);
    void SetParams(double start, double end, double cycle, bool update = false);
    void ShouldCycle(bool sc) { m_shouldCycle = sc; }

src/common/engine/d_event.cpp

@@ -67,11 +67,21 @@ CVAR(Bool, m_filter, false, CVAR_ARCHIVE | CVAR_GLOBALCONFIG)
 
 void D_ProcessEvents (void)
 {
-	event_t *ev;
+	FixedBitArray<NUM_KEYS> keywasdown;
+	TArray<event_t> delayedevents;
+
+	keywasdown.Zero();
 	while (eventtail != eventhead)
 	{
-		ev = &events[eventtail];
+		event_t *ev = &events[eventtail];
 		eventtail = (eventtail + 1) & (MAXEVENTS - 1);
+
+		if (ev->type == EV_KeyUp && keywasdown[ev->data1])
+		{
+			delayedevents.Push(*ev);
+			continue;
+		}
+
 		if (ev->type == EV_None)
 			continue;
 		if (ev->type == EV_DeviceChange)
@@ -85,7 +95,12 @@ void D_ProcessEvents (void)
 				continue;				// menu ate the event
 		}
 
-		G_Responder (ev);
+		if (G_Responder(ev) && ev->type == EV_KeyDown) keywasdown.Set(ev->data1);
+	}
+
+	for (auto& ev: delayedevents)
+	{
+		D_PostEvent(&ev);
 	}
 }
 
@@ -123,7 +138,7 @@ void D_RemoveNextCharEvent()
 		}
 	}
 }
- 
+
 
 //==========================================================================
 //

src/common/engine/i_net.cpp

@@ -7,7 +7,7 @@
 //
 // This program is free software: you can redistribute it and/or modify
 // it under the terms of the GNU General Public License as published by
-// the Free Software Foundation, either version 3 of the License, or
+// the Free Software Foundation, either version 2 of the License, or
 // (at your option) any later version.
 //
 // This program is distributed in the hope that it will be useful,
@@ -172,7 +172,7 @@ FString GetPlayerName(int num)
 SOCKET UDPsocket (void)
 {
 	SOCKET s;
-		
+
 	// allocate a socket
 	s = socket (PF_INET, SOCK_DGRAM, IPPROTO_UDP);
 	if (s == INVALID_SOCKET)
@@ -193,7 +193,7 @@ void BindToLocalPort (SOCKET s, u_short port)
 	address.sin_family = AF_INET;
 	address.sin_addr.s_addr = INADDR_ANY;
 	address.sin_port = htons(port);
-						
+
 	v = bind (s, (sockaddr *)&address, sizeof(address));
 	if (v == SOCKET_ERROR)
 		I_FatalError ("BindToPort: %s", neterror ());
@@ -459,7 +459,7 @@ void StartNetwork (bool autoPort)
 
 	netgame = true;
 	multiplayer = true;
-	
+
 	// create communication socket
 	mysocket = UDPsocket ();
 	BindToLocalPort (mysocket, autoPort ? 0 : DOOMPORT);
@@ -736,7 +736,7 @@ bool HostGame (int i)
 	{
 		// If we send the packets eight times to each guest,
 		// hopefully at least one of them will get through.
-		for (int i = 8; i != 0; --i)
+		for (int ii = 8; ii != 0; --ii)
 		{
 			PreSend (&packet, 2, &sendaddress[node]);
 		}
@@ -892,7 +892,7 @@ bool JoinGame (int i)
 		SendAbort();
 		return false;
 	}
-	
+
 	StartScreen->NetMessage ("Total players: %d", doomcom.numnodes);
 
 	doomcom.id = DOOMCOM_ID;

src/common/engine/i_net.h

@@ -76,7 +76,7 @@ struct doomcom_t
 
 // packet data to be sent
 	uint8_t	data[MAX_MSGLEN];
-	
+
 }; 
 
 extern doomcom_t doomcom;

src/common/engine/m_joy.h

@@ -21,7 +21,7 @@ enum EJoyAxis
 struct NOVTABLE IJoystickConfig
 {
 	virtual ~IJoystickConfig() = 0;
-	
+
 	virtual FString GetName() = 0;
 	virtual float GetSensitivity() = 0;
 	virtual void SetSensitivity(float scale) = 0;

src/common/engine/palettecontainer.cpp

@@ -234,7 +234,7 @@ void PaletteContainer::UpdateTranslation(int trans, FRemapTable* remap)
 
 int PaletteContainer::AddTranslation(int slot, FRemapTable* remap, int count)
 {
-	uint32_t id;
+	uint32_t id = 0;
 	for (int i = 0; i < count; i++)
 	{
 		auto newremap = AddRemap(&remap[i]);
@@ -265,7 +265,7 @@ FRemapTable *PaletteContainer::TranslationToTable(int translation)
 	unsigned int type = GetTranslationType(translation);
 	unsigned int index = GetTranslationIndex(translation);
 
-	if (type < 0 || type >= TranslationTables.Size() || index >= NumTranslations(type))
+	if (type >= TranslationTables.Size() || index >= NumTranslations(type))
 	{
 		return uniqueRemaps[0]; // this is the identity table.
 	}
@@ -649,7 +649,6 @@ bool FRemapTable::AddTint(int start, int end, int r, int g, int b, int amount)
 bool FRemapTable::AddToTranslation(const char *range)
 {
 	int start,end;
-	bool desaturated = false;
 	FScanner sc;
 
 	sc.OpenMem("translation", range, int(strlen(range)));

src/common/engine/sc_man.cpp

@@ -857,7 +857,7 @@ bool FScanner::CheckFloat (bool evaluate)
 			UnGet();
 			return false;
 		}
-	
+
 		Float = strtod (String, &stopper);
 		if (*stopper != 0)
 		{
@@ -1131,9 +1131,7 @@ FString FScanner::TokenName (int token, const char *string)
 	}
 	else
 	{
-		FString work;
+		work.Format("Unknown(%d)", token);
-		work.Format ("Unknown(%d)", token);
-		return work;
 	}
 	return work;
 }

src/common/engine/sc_man.h

@@ -159,7 +159,7 @@ public:
 	{
 		return constants.CheckKey(name);
 	}
-	
+
 	// Token based variant
 	bool CheckValue(bool allowfloat, bool evaluate = true);
 	void MustGetValue(bool allowfloat, bool evaluate = true);

src/common/engine/sc_man_scanner.re

@@ -102,14 +102,14 @@ std2:
 		":"							{ RET(':'); }
 		";"							{ RET(';'); }
 		"}"							{ StateMode = 0; StateOptions = false; RET('}'); }
-		
+
 		WSP+						{ goto std1; }
 		"\n"						{ goto newline; }
-		
+
 		TOKS = (NWS\[/":;}]);
 		TOKS* ([/] (TOKS\[*]) TOKS*)*
 									{ RET(TK_NonWhitespace); }
-		
+
 	*/
 	}
 	else if (tokens)	// A well-defined scanner, based on the c.re example.
@@ -222,7 +222,7 @@ std2:
 		'canraise'					{ RET(StateOptions ? TK_CanRaise : TK_Identifier); }
 		'offset'					{ RET(StateOptions ? TK_Offset : TK_Identifier); }
 		'light'						{ RET(StateOptions ? TK_Light : TK_Identifier); }
-		
+
 		/* other DECORATE top level keywords */
 		'#include'					{ RET(TK_Include); }
 

src/common/engine/serializer.cpp

@@ -290,6 +290,28 @@ bool FSerializer::BeginObject(const char *name)
 //
 //==========================================================================
 
+bool FSerializer::HasObject(const char* name)
+{
+	if (isReading())
+	{
+		auto val = r->FindKey(name);
+		if (val != nullptr)
+		{
+			if (val->IsObject())
+			{
+				return true;
+			}
+		}
+	}
+	return false;
+}
+
+//==========================================================================
+//
+//
+//
+//==========================================================================
+
 void FSerializer::EndObject()
 {
 	if (isWriting())
@@ -601,7 +623,7 @@ void FSerializer::WriteObjects()
 void FSerializer::ReadObjects(bool hubtravel)
 {
 	bool founderrors = false;
-	
+
 	if (isReading() && BeginArray("objects"))
 	{
 		// Do not link any thinker that's being created here. This will be done by deserializing the thinker list later.
@@ -619,7 +641,6 @@ void FSerializer::ReadObjects(bool hubtravel)
 				if (BeginObject(nullptr))
 				{
 					FString clsname;	// do not deserialize the class type directly so that we can print appropriate errors.
-					int pindex = -1;
 
 					Serialize(*this, "classtype", clsname, nullptr);
 					PClass *cls = PClass::FindClass(clsname);
@@ -643,6 +664,7 @@ void FSerializer::ReadObjects(bool hubtravel)
 			if (!founderrors)
 			{
 				// Reset to start;
+				unsigned size = r->mObjects.Size();
 				r->mObjects.Last().mIndex = 0;
 
 				for (unsigned i = 0; i < r->mDObjects.Size(); i++)
@@ -652,7 +674,6 @@ void FSerializer::ReadObjects(bool hubtravel)
 					{
 						if (obj != nullptr)
 						{
-							int pindex = -1;
 							try
 							{
 								obj->SerializeUserVars(*this);
@@ -660,6 +681,7 @@ void FSerializer::ReadObjects(bool hubtravel)
 							}
 							catch (CRecoverableError &err)
 							{
+								r->mObjects.Clamp(size);	// close all inner objects.
 								// In case something in here throws an error, let's continue and deal with it later.
 								Printf(TEXTCOLOR_RED "'%s'\n while restoring %s\n", err.GetMessage(), obj ? obj->GetClass()->TypeName.GetChars() : "invalid object");
 								mErrors++;

src/common/engine/serializer.h

@@ -85,6 +85,7 @@ public:
 	void ReadObjects(bool hubtravel);
 	bool BeginObject(const char *name);
 	void EndObject();
+	bool HasObject(const char* name);
 	bool BeginArray(const char *name);
 	void EndArray();
 	unsigned GetSize(const char *group);
@@ -110,7 +111,7 @@ public:
 	{
 		return w != nullptr;
 	}
-	
+
 	bool canSkip() const;
 
 	template<class T>
@@ -234,7 +235,7 @@ FSerializer &Serialize(FSerializer &arc, const char *key, FSoundID &sid, FSoundI
 FSerializer &Serialize(FSerializer &arc, const char *key, FString &sid, FString *def);
 FSerializer &Serialize(FSerializer &arc, const char *key, NumericValue &sid, NumericValue *def);
 
-template<class T>
+template <typename T/*, typename = std::enable_if_t<std::is_base_of_v<DObject, T>>*/>
 FSerializer &Serialize(FSerializer &arc, const char *key, T *&value, T **)
 {
 	DObject *v = static_cast<DObject*>(value);
@@ -301,6 +302,11 @@ FSerializer& Serialize(FSerializer& arc, const char* key, FixedBitArray<size>& v
 	return arc.SerializeMemory(key, value.Storage(), value.StorageSize());
 }
 
+inline FSerializer& Serialize(FSerializer& arc, const char* key, BitArray& value, BitArray* def)
+{
+	return arc.SerializeMemory(key, value.Storage().Data(), value.Storage().Size());
+}
+
 template<> FSerializer& Serialize(FSerializer& arc, const char* key, PClass*& clst, PClass** def);
 template<> FSerializer& Serialize(FSerializer& arc, const char* key, FFont*& font, FFont** def);
 template<> FSerializer &Serialize(FSerializer &arc, const char *key, Dictionary *&dict, Dictionary **def);

src/common/engine/serializer_internal.h

@@ -42,7 +42,7 @@ struct FWriter
 	rapidjson::StringBuffer mOutString;
 	TArray<DObject *> mDObjects;
 	TMap<DObject *, int> mObjectMap;
-	
+
 	FWriter(bool pretty)
 	{
 		if (!pretty)
@@ -193,7 +193,7 @@ struct FReader
 	rapidjson::Value *FindKey(const char *key)
 	{
 		FJSONObject &obj = mObjects.Last();
-		
+
 		if (obj.mObject->IsObject())
 		{
 			if (key == nullptr)

src/common/engine/st_start.h

@@ -47,7 +47,7 @@ public:
 		CurPos = 0;
 		NotchPos = 0;
 	}
-	
+
 	virtual ~FStartupScreen() = default;
 
 	virtual void Progress() {}
@@ -98,7 +98,7 @@ public:
 	void AppendStatusLine(const char *status);
 protected:
 	void SetWindowSize();
-	
+
 	int ThermX, ThermY, ThermWidth, ThermHeight;
 	int HMsgY, SMsgX;
 };

src/common/engine/stats.h

@@ -61,28 +61,28 @@ public:
 	{
 		Sec = 0;
 	}
-	
+
 	void Clock()
 	{
 		timespec ts;
-		
+
 		clock_gettime(CLOCK_MONOTONIC, &ts);
 		Sec -= ts.tv_sec + ts.tv_nsec * 1e-9;
 	}
-	
+
 	void Unclock()
 	{
 		timespec ts;
-		
+
 		clock_gettime(CLOCK_MONOTONIC, &ts);
 		Sec += ts.tv_sec + ts.tv_nsec * 1e-9;
 	}
-	
+
 	double Time()
 	{
 		return Sec;
 	}
-	
+
 	double TimeMS()
 	{
 		return Sec * 1e3;
@@ -159,24 +159,24 @@ public:
 	{
 		Counter = 0;
 	}
-	
+
 	void Clock()
 	{
 		int64_t time = rdtsc();
 		Counter -= time;
 	}
-	
+
 	void Unclock(bool checkvar = true)
 	{
 		int64_t time = rdtsc();
 		Counter += time;
 	}
-	
+
 	double Time()
 	{
 		return Counter * PerfToSec;
 	}
-	
+
 	double TimeMS()
 	{
 		return Counter * PerfToMillisec;

src/common/engine/stringtable.cpp

@@ -639,7 +639,7 @@ bool FStringTable::MatchDefaultString(const char *name, const char *content) con
 	// This only compares the first line to avoid problems with bad linefeeds. For the few cases where this feature is needed it is sufficient.
 	auto c = GetLanguageString(name, FStringTable::default_table);
 	if (!c) return false;
-	
+
 	// Check a secondary key, in case the text comparison cannot be done due to needed orthographic fixes (see Harmony's exit text)
 	FStringf checkkey("%s_CHECK", name);
 	auto cc = GetLanguageString(checkkey, FStringTable::default_table);

src/common/engine/stringtable.h

@@ -91,7 +91,7 @@ public:
 		allStrings.Insert(override_table, map);
 		UpdateLanguage(nullptr);
 	}
-	
+
 	const char *GetLanguageString(const char *name, uint32_t langtable, int gender = -1) const;
 	bool MatchDefaultString(const char *name, const char *content) const;
 	const char *GetString(const char *name, uint32_t *langtable, int gender = -1) const;
@@ -110,7 +110,7 @@ private:
 	StringMacroMap allMacros;
 	LangMap allStrings;
 	TArray<std::pair<uint32_t, StringMap*>> currentLanguageSet;
-	
+
 	void LoadLanguage (int lumpnum, const TArray<uint8_t> &buffer);
 	TArray<TArray<FString>> parseCSV(const TArray<uint8_t> &buffer);
 	bool ParseLanguageCSV(int lumpnum, const TArray<uint8_t> &buffer);
@@ -127,7 +127,7 @@ public:
 		if (*str == '$') return str;
 		return FString("$") + str;
 	}
-	
+
 	static FString MakeMacro(const char *str, size_t len)
 	{
 		if (*str == '$') return FString(str, len);

src/common/engine/v_colortables.cpp

@@ -57,7 +57,7 @@ ColorTable256k RGB256k;
 void BuildTransTable (const PalEntry *palette)
 {
 	int r, g, b;
-	
+
 	// create the RGB555 lookup table
 	for (r = 0; r < 32; r++)
 		for (g = 0; g < 32; g++)
@@ -68,16 +68,16 @@ void BuildTransTable (const PalEntry *palette)
 		for (g = 0; g < 64; g++)
 			for (b = 0; b < 64; b++)
 				RGB256k.RGB[r][g][b] = ColorMatcher.Pick ((r<<2)|(r>>4), (g<<2)|(g>>4), (b<<2)|(b>>4));
-	
+
 	int x, y;
-	
+
 	// create the swizzled palette
 	for (x = 0; x < 65; x++)
 		for (y = 0; y < 256; y++)
 			Col2RGB8[x][y] = (((palette[y].r*x)>>4)<<20) |
 			((palette[y].g*x)>>4) |
 			(((palette[y].b*x)>>4)<<10);
-	
+
 	// create the swizzled palette with the lsb of red and blue forced to 0
 	// (for green, a 1 is okay since it never gets added into)
 	for (x = 1; x < 64; x++)
@@ -90,7 +90,7 @@ void BuildTransTable (const PalEntry *palette)
 	}
 	Col2RGB8_LessPrecision[0] = Col2RGB8[0];
 	Col2RGB8_LessPrecision[64] = Col2RGB8[64];
-	
+
 	// create the inverse swizzled palette
 	for (x = 0; x < 65; x++)
 		for (y = 0; y < 256; y++)

src/common/filesystem/file_directory.cpp

@@ -41,8 +41,6 @@
 #include "printf.h"
 #include "findfile.h"
 
-
-
 //==========================================================================
 //
 // Zip Lump
@@ -120,7 +118,7 @@ int FDirectory::AddDirectory(const char *dirpath)
 	FString dirmatch = dirpath;
 	findstate_t find;
 	dirmatch += '*';
-	
+
 	handle = I_FindFirst(dirmatch.GetChars(), &find);
 	if (handle == ((void *)(-1)))
 	{

src/common/filesystem/file_grp.cpp

@@ -99,7 +99,7 @@ bool FGrpFile::Open(bool quiet, LumpFilterInfo*)
 
 	Reader.Read(&header, sizeof(header));
 	NumLumps = LittleLong(header.NumLumps);
-	
+
 	GrpLump *fileinfo = new GrpLump[NumLumps];
 	Reader.Read (fileinfo, NumLumps * sizeof(GrpLump));
 

src/common/filesystem/file_pak.cpp

@@ -95,7 +95,7 @@ bool FPakFile::Open(bool quiet, LumpFilterInfo* filter)
 	Reader.Read(&header, sizeof(header));
 	NumLumps = LittleLong(header.dirlen) / sizeof(dpackfile_t);
 	header.dirofs = LittleLong(header.dirofs);
-	
+
 	TArray<dpackfile_t> fileinfo(NumLumps, true);
 	Reader.Seek (header.dirofs, FileReader::SeekSet);
 	Reader.Read (fileinfo.Data(), NumLumps * sizeof(dpackfile_t));

src/common/filesystem/file_rff.cpp

@@ -221,7 +221,7 @@ int FRFFLump::FillCache()
 	{
 		int cryptlen = min<int> (LumpSize, 256);
 		uint8_t *data = (uint8_t *)Cache;
-		
+
 		for (int i = 0; i < cryptlen; ++i)
 		{
 			data[i] ^= i >> 1;

src/common/filesystem/file_wad.cpp

@@ -206,7 +206,7 @@ bool FWadFile::Open(bool quiet, LumpFilterInfo*)
 		Lumps[i].LumpSize = isBigEndian ? BigLong(fileinfo[i].Size) : LittleLong(fileinfo[i].Size);
 		Lumps[i].Namespace = ns_global;
 		Lumps[i].Flags = Lumps[i].Compressed ? LUMPF_COMPRESSED | LUMPF_SHORTNAME : LUMPF_SHORTNAME;
-		
+
 		// Check if the lump is within the WAD file and print a warning if not.
 		if (Lumps[i].Position + Lumps[i].LumpSize > wadSize || Lumps[i].Position < 0 || Lumps[i].LumpSize < 0)
 		{
@@ -279,7 +279,7 @@ void FWadFile::SetNamespace(const char *startmarker, const char *endmarker, name
 	int numstartmarkers = 0, numendmarkers = 0;
 	unsigned int i;
 	TArray<Marker> markers;
-	
+
 	for(i = 0; i < NumLumps; i++)
 	{
 		if (IsMarker(i, startmarker))
@@ -302,20 +302,20 @@ void FWadFile::SetNamespace(const char *startmarker, const char *endmarker, name
 
 		Printf(TEXTCOLOR_YELLOW"WARNING: %s marker without corresponding %s found.\n", endmarker, startmarker);
 
-		
+
 		if (flathack)
 		{
 			// We have found no F_START but one or more F_END markers.
 			// mark all lumps before the last F_END marker as potential flats.
 			unsigned int end = markers[markers.Size()-1].index;
-			for(unsigned int i = 0; i < end; i++)
+			for(unsigned int ii = 0; ii < end; ii++)
 			{
-				if (Lumps[i].LumpSize == 4096)
+				if (Lumps[ii].LumpSize == 4096)
 				{
 					// We can't add this to the flats namespace but 
 					// it needs to be flagged for the texture manager.
-					DPrintf(DMSG_NOTIFY, "Marking %s as potential flat\n", Lumps[i].getName());
+					DPrintf(DMSG_NOTIFY, "Marking %s as potential flat\n", Lumps[ii].getName());
-					Lumps[i].Flags |= LUMPF_MAYBEFLAT;
+					Lumps[ii].Flags |= LUMPF_MAYBEFLAT;
 				}
 			}
 		}
@@ -428,18 +428,19 @@ void FWadFile::SkinHack ()
 				namespc++;
 			}
 		}
+		// needless to say, this check is entirely useless these days as map names can be more diverse..
 		if ((lump->getName()[0] == 'M' &&
 			 lump->getName()[1] == 'A' &&
 			 lump->getName()[2] == 'P' &&
 			 lump->getName()[3] >= '0' && lump->getName()[3] <= '9' &&
 			 lump->getName()[4] >= '0' && lump->getName()[4] <= '9' &&
-			 lump->getName()[5] >= '\0')
+			 lump->getName()[5] == '\0')
 			||
 			(lump->getName()[0] == 'E' &&
 			 lump->getName()[1] >= '0' && lump->getName()[1] <= '9' &&
 			 lump->getName()[2] == 'M' &&
 			 lump->getName()[3] >= '0' && lump->getName()[3] <= '9' &&
-			 lump->getName()[4] >= '\0'))
+			 lump->getName()[4] == '\0'))
 		{
 			hasmap = true;
 		}

src/common/filesystem/file_whres.cpp

@@ -95,10 +95,10 @@ FWHResFile::FWHResFile(const char *filename, FileReader &file)
 bool FWHResFile::Open(bool quiet, LumpFilterInfo*)
 {
 	int directory[1024];
-	
+
 	Reader.Seek(-4096, FileReader::SeekEnd);
 	Reader.Read(directory, 4096);
-	
+
 	int nl =1024/3;
 	Lumps.Resize(nl);
 
@@ -135,10 +135,10 @@ FResourceFile *CheckWHRes(const char *filename, FileReader &file, bool quiet, Lu
 	{
 		int directory[1024];
 		int nl =1024/3;
-		
+
 		file.Seek(-4096, FileReader::SeekEnd);
 		file.Read(directory, 4096);
-		
+
 		int checkpos = 0;
 		for(int k = 0; k < nl; k++)
 		{

src/common/filesystem/filesystem.cpp

@@ -218,14 +218,13 @@ void FileSystem::InitMultipleFiles (TArray<FString> &filenames, bool quiet, Lump
 
 	for(unsigned i=0;i<filenames.Size(); i++)
 	{
-		int baselump = NumEntries;
 		AddFile (filenames[i], nullptr, quiet, filter);
-		
+
 		if (i == (unsigned)MaxIwadIndex) MoveLumpsInFolder("after_iwad/");
 		FStringf path("filter/%s", Files.Last()->GetHash().GetChars());
 		MoveLumpsInFolder(path);
 	}
-	
+
 	NumEntries = FileInfo.Size();
 	if (NumEntries == 0)
 	{
@@ -284,7 +283,7 @@ int FileSystem::AddFromBuffer(const char* name, const char* type, char* data, in
 	FileInfo.Last().resourceId = id;
 	return FileInfo.Size()-1;
 }
- 
+
 //==========================================================================
 //
 // AddFile
@@ -333,7 +332,7 @@ void FileSystem::AddFile (const char *filename, FileReader *filer, bool quiet, L
 	startlump = NumEntries;
 
 	FResourceFile *resfile;
-	
+
 	if (!isdir)
 		resfile = FResourceFile::OpenResourceFile(filename, filereader, quiet, false, filter);
 	else
@@ -946,10 +945,10 @@ void FileSystem::MoveLumpsInFolder(const char *path)
 	{
 		return;
 	}
-	
+
 	auto len = strlen(path);
 	auto rfnum = FileInfo.Last().rfnum;
-	
+
 	unsigned i;
 	for (i = 0; i < FileInfo.Size(); i++)
 	{
@@ -1022,7 +1021,7 @@ int FileSystem::FindLumpMulti (const char **names, int *lastlump, bool anyns, in
 	{
 		if (anyns || lump_p->Namespace == ns_global)
 		{
-			
+
 			for(const char **name = names; *name != NULL; name++)
 			{
 				if (!strnicmp(*name, lump_p->shortName.String, 8))
@@ -1512,7 +1511,7 @@ int FileSystem::GetEntryCount (int rfnum) const noexcept
 	{
 		return 0;
 	}
-	
+
 	return Files[rfnum]->LumpCount();
 }
 
@@ -1669,3 +1668,20 @@ FResourceLump* FileSystem::GetFileAt(int no)
 	return FileInfo[no].lump;
 }
 
+#include "c_dispatch.h"
+
+CCMD(fs_dir)
+{
+	int numfiles = fileSystem.GetNumEntries();
+
+	for (int i = 0; i < numfiles; i++)
+	{
+		auto container = fileSystem.GetResourceFileFullName(fileSystem.GetFileContainer(i));
+		auto fn1 = fileSystem.GetFileFullName(i);
+		auto fns = fileSystem.GetFileShortName(i);
+		auto fnid = fileSystem.GetResourceId(i);
+		auto length = fileSystem.FileLength(i);
+		bool hidden = fileSystem.FindFile(fn1) != i;
+		Printf(PRINT_NONOTIFY, "%s%-64s %-15s (%5d) %10d %s %s\n", hidden ? TEXTCOLOR_RED : TEXTCOLOR_UNTRANSLATED, fn1, fns, fnid, length, container, hidden ? "(h)" : "");
+	}
+}

src/common/filesystem/resourcefile.cpp

@@ -306,11 +306,11 @@ int lumpcmp(const void * a, const void * b)
 void FResourceFile::GenerateHash()
 {
 	// hash the lump directory after sorting
-	
+
 	Hash.Format(("%08X-%04X-"), (unsigned)Reader.GetLength(), NumLumps);
-	
+
 	MD5Context md5;
-	
+
 	uint8_t digest[16];
 	for(uint32_t i = 0; i < NumLumps; i++)
 	{
@@ -382,9 +382,9 @@ int FResourceFile::FilterLumps(FString filtername, void *lumps, size_t lumpsize,
 		return 0;
 	}
 	filter << "filter/" << filtername << '/';
-	
+
 	bool found = FindPrefixRange(filter, lumps, lumpsize, max, start, end);
-	
+
 	// Workaround for old Doom filter names.
 	if (!found && filtername.IndexOf("doom.id.doom") == 0)
 	{
@@ -490,7 +490,7 @@ bool FResourceFile::FindPrefixRange(FString filter, void *lumps, size_t lumpsize
 {
 	uint32_t min, max, mid, inside;
 	FResourceLump *lump;
-	int cmp;
+	int cmp = 0;
 
 	end = start = 0;
 
@@ -499,7 +499,7 @@ bool FResourceFile::FindPrefixRange(FString filter, void *lumps, size_t lumpsize
 	lumps = (uint8_t *)lumps - lumpsize;
 
 	// Binary search to find any match at all.
-	min = 1, max = maxlump;
+	mid = min = 1, max = maxlump;
 	while (min <= max)
 	{
 		mid = min + (max - min) / 2;

src/common/filesystem/resourcefile.h

@@ -6,6 +6,7 @@
 #include <limits.h>
 
 #include "files.h"
+#include "zstring.h"
 
 struct LumpFilterInfo
 {

src/common/fonts/font.cpp

@@ -86,10 +86,10 @@ FFont::FFont (const char *name, const char *nametemplate, const char *filetempla
 	TMap<int, FGameTexture*> charMap;
 	int minchar = INT_MAX;
 	int maxchar = INT_MIN;
-	
+
 	// Read the font's configuration.
 	// This will not be done for the default fonts, because they are not atomic and the default content does not need it.
-	
+
 	TArray<FolderEntry> folderdata;
 	if (filetemplate != nullptr)
 	{
@@ -97,16 +97,16 @@ FFont::FFont (const char *name, const char *nametemplate, const char *filetempla
 		// If a name template is given, collect data from all resource files.
 		// For anything else, each folder is being treated as an atomic, self-contained unit and mixing from different glyph sets is blocked.
 		fileSystem.GetFilesInFolder(path, folderdata, nametemplate == nullptr);
-		
+
 		//if (nametemplate == nullptr)
 		{
 			FStringf infpath("fonts/%s/font.inf", filetemplate);
-			
+
 			unsigned index = folderdata.FindEx([=](const FolderEntry &entry)
 			{
 				return infpath.CompareNoCase(entry.name) == 0;
 			});
-			
+
 			if (index < folderdata.Size())
 			{
 				FScanner sc;
@@ -182,7 +182,7 @@ FFont::FFont (const char *name, const char *nametemplate, const char *filetempla
 			}
 		}
 	}
-	
+
 	if (FixedWidth > 0)
 	{
 		ReadSheetFont(folderdata, FixedWidth, FontHeight, Scale);
@@ -292,12 +292,12 @@ FFont::FFont (const char *name, const char *nametemplate, const char *filetempla
 				auto position = strtoll(base.GetChars(), &endp, 16);
 				if ((*endp == 0 || (*endp == '.' && position >= '!' && position < 0xffff)))
 				{
-					auto lump = TexMan.CheckForTexture(entry.name, ETextureType::MiscPatch);
+					auto texlump = TexMan.CheckForTexture(entry.name, ETextureType::MiscPatch);
-					if (lump.isValid())
+					if (texlump.isValid())
 					{
 						if ((int)position < minchar) minchar = (int)position;
 						if ((int)position > maxchar) maxchar = (int)position;
-						auto tex = TexMan.GetGameTexture(lump);
+						auto tex = TexMan.GetGameTexture(texlump);
 						tex->SetScale((float)Scale.X, (float)Scale.Y);
 						charMap.Insert((int)position, tex);
 						Type = Folder;
@@ -313,10 +313,10 @@ FFont::FFont (const char *name, const char *nametemplate, const char *filetempla
 
 		for (i = 0; i < count; i++)
 		{
-			auto lump = charMap.CheckKey(FirstChar + i);
+			auto charlump = charMap.CheckKey(FirstChar + i);
-			if (lump != nullptr)
+			if (charlump != nullptr)
 			{
-				auto pic = *lump;
+				auto pic = *charlump;
 				if (pic != nullptr)
 				{
 					double fheight = pic->GetDisplayHeight();
@@ -399,8 +399,8 @@ void FFont::ReadSheetFont(TArray<FolderEntry> &folderdata, int width, int height
 						part[0].OriginY = -height * y;
 						part[0].TexImage = static_cast<FImageTexture*>(tex->GetTexture());
 						FMultiPatchTexture *image = new FMultiPatchTexture(width, height, part, false, false);
-						FImageTexture *tex = new FImageTexture(image);
+						FImageTexture *imgtex = new FImageTexture(image);
-						auto gtex = MakeGameTexture(tex, nullptr, ETextureType::FontChar);
+						auto gtex = MakeGameTexture(imgtex, nullptr, ETextureType::FontChar);
 						gtex->SetWorldPanning(true);
 						gtex->SetOffsets(0, 0, 0);
 						gtex->SetOffsets(1, 0, 0);
@@ -424,7 +424,6 @@ void FFont::ReadSheetFont(TArray<FolderEntry> &folderdata, int width, int height
 	LastChar = maxchar;
 	auto count = maxchar - minchar + 1;
 	Chars.Resize(count);
-	int fontheight = 0;
 
 	for (int i = 0; i < count; i++)
 	{
@@ -560,10 +559,10 @@ FFont *FFont::FindFont (FName name)
 void RecordTextureColors (FImageSource *pic, uint32_t *usedcolors)
 {
 	int x;
-	
+
 	auto pixels = pic->GetPalettedPixels(false);
 	auto size = pic->GetWidth() * pic->GetHeight();
-	
+
 	for(x = 0;x < size; x++)
 	{
 		usedcolors[pixels[x]]++;
@@ -728,7 +727,7 @@ int FFont::GetCharCode(int code, bool needpic) const
 	{
 		return code;
 	}
-	
+
 	// Use different substitution logic based on the fonts content:
 	// In a font which has both upper and lower case, prefer unaccented small characters over capital ones.
 	// In a pure upper-case font, do not check for lower case replacements.
@@ -806,7 +805,7 @@ FGameTexture *FFont::GetChar (int code, int translation, int *const width) const
 		code -= FirstChar;
 		xmove = Chars[code].XMove;
 	}
-	
+
 	if (width != nullptr)
 	{
 		*width = xmove;

src/common/fonts/hexfont.cpp

@@ -266,7 +266,7 @@ int FHexFontChar2::CopyPixels(FBitmap* bmp, int conversion)
 
 class FHexFont : public FFont
 {
-	
+
 public:
 	//==========================================================================
 	//
@@ -283,14 +283,14 @@ public:
 		assert(lump >= 0);
 
 		FontName = fontname;
-		
+
 		FirstChar = hexdata.FirstChar;
 		LastChar = hexdata.LastChar;
 
 		FontHeight = 16;
 		SpaceWidth = 9;
 		GlobalKerning = 0;
-		
+
 		Chars.Resize(LastChar - FirstChar + 1);
 		for (int i = FirstChar; i <= LastChar; i++)
 		{
@@ -325,7 +325,7 @@ public:
 			else Translations[i] = LuminosityTranslation(i * 2 + 1, minlum, maxlum);
 		}
 	}
-	
+
 };
 
 

src/common/fonts/singlelumpfont.cpp

@@ -96,7 +96,7 @@ protected:
 	void LoadFON1 (int lump, const uint8_t *data);
 	void LoadFON2 (int lump, const uint8_t *data);
 	void LoadBMF (int lump, const uint8_t *data);
-	
+
 	enum
 	{
 		FONT1,
@@ -259,7 +259,7 @@ void FSingleLumpFont::LoadFON2 (int lump, const uint8_t *data)
 	LastChar = data[7];
 	ActiveColors = data[10]+1;
 	RescalePalette = data[9] == 0;
-	
+
 	count = LastChar - FirstChar + 1;
 	Chars.Resize(count);
 	TArray<int> widths2(count, true);
@@ -309,9 +309,9 @@ void FSingleLumpFont::LoadFON2 (int lump, const uint8_t *data)
 	}
 
 	Palette[0] = 0;
-	for (int i = 1; i < ActiveColors; i++)
+	for (int pp = 1; pp < ActiveColors; pp++)
 	{
-		Palette[i] = PalEntry(255, palette[i * 3], palette[i * 3 + 1], palette[i * 3 + 2]);
+		Palette[pp] = PalEntry(255, palette[pp * 3], palette[pp * 3 + 1], palette[pp * 3 + 2]);
 	}
 
 	data_p = palette + ActiveColors*3;

src/common/fonts/specialfont.cpp

@@ -102,12 +102,12 @@ FSpecialFont::FSpecialFont (const char *name, int first, int count, FGameTexture
 
 		if (charlumps[i] != nullptr)
 		{
-			auto pic = charlumps[i];
+			auto charpic = charlumps[i];
-			Chars[i].OriginalPic = MakeGameTexture(pic->GetTexture(), nullptr, ETextureType::FontChar);
+			Chars[i].OriginalPic = MakeGameTexture(charpic->GetTexture(), nullptr, ETextureType::FontChar);
-			Chars[i].OriginalPic->CopySize(pic, true);
+			Chars[i].OriginalPic->CopySize(charpic, true);
 			TexMan.AddGameTexture(Chars[i].OriginalPic);
 			Chars[i].XMove = (int)Chars[i].OriginalPic->GetDisplayWidth();
-			if (sysCallbacks.FontCharCreated) sysCallbacks.FontCharCreated(pic, Chars[i].OriginalPic);
+			if (sysCallbacks.FontCharCreated) sysCallbacks.FontCharCreated(charpic, Chars[i].OriginalPic);
 		}
 		else
 		{

src/common/fonts/v_font.cpp

@@ -106,10 +106,10 @@ FFont *V_GetFont(const char *name, const char *fontlumpname)
 
 		int lump = -1;
 		int folderfile = -1;
-		
+
 		TArray<FolderEntry> folderdata;
 		FStringf path("fonts/%s/", name);
-		
+
 		// Use a folder-based font only if it comes from a later file than the single lump version.
 		if (fileSystem.GetFilesInFolder(path, folderdata, true))
 		{
@@ -119,7 +119,7 @@ FFont *V_GetFont(const char *name, const char *fontlumpname)
 
 
 		lump = fileSystem.CheckNumForFullName(fontlumpname? fontlumpname : name, true);
-		
+
 		if (lump != -1 && fileSystem.GetFileContainer(lump) >= folderfile)
 		{
 			uint32_t head;
@@ -725,9 +725,9 @@ static void CalcDefaultTranslation(FFont* base, int index)
 		auto lum = otherluminosity[i];
 		if (lum >= 0 && lum <= 1)
 		{
-			int index = int(lum * 255);
+			int lumidx = int(lum * 255);
-			remap[index] = GPalette.BaseColors[i];
+			remap[lumidx] = GPalette.BaseColors[i];
-			remap[index].a = 255;
+			remap[lumidx].a = 255;
 		}
 	}
 
@@ -769,7 +769,7 @@ static void CalcDefaultTranslation(FFont* base, int index)
 			lowindex = highindex++;
 		}
 	}
-	
+
 }
 
 //==========================================================================

src/common/fonts/v_text.cpp

@@ -88,7 +88,6 @@ TArray<FBrokenLines> V_BreakLines (FFont *font, int maxwidth, const uint8_t *str
 			{
 				if (*string == '[')
 				{
-					const uint8_t* start = string;
 					while (*string != ']' && *string != '\0')
 					{
 						string++;

src/common/menu/joystickmenu.cpp

@@ -160,12 +160,12 @@ void UpdateJoystickMenu(IJoystickConfig *selected)
 		it = opt->GetItem("ConnectMessage2");
 		if (it != nullptr) it->SetValue(0, !use_joystick);
 
-		for (int i = 0; i < (int)Joysticks.Size(); ++i)
+		for (int ii = 0; ii < (int)Joysticks.Size(); ++ii)
 		{
-			it = CreateOptionMenuItemJoyConfigMenu(Joysticks[i]->GetName(), Joysticks[i]);
+			it = CreateOptionMenuItemJoyConfigMenu(Joysticks[ii]->GetName(), Joysticks[ii]);
 			GC::WriteBarrier(opt, it);
 			opt->mItems.Push(it);
-			if (i == itemnum) opt->mSelectedItem = opt->mItems.Size();
+			if (ii == itemnum) opt->mSelectedItem = opt->mItems.Size();
 		}
 		if (opt->mSelectedItem >= (int)opt->mItems.Size())
 		{
@@ -179,15 +179,15 @@ void UpdateJoystickMenu(IJoystickConfig *selected)
 			auto p = CurrentMenu->PointerVar<IJoystickConfig>("mJoy");
 			if (p != nullptr)
 			{
-				unsigned i;
+				unsigned ii;
-				for (i = 0; i < Joysticks.Size(); ++i)
+				for (ii = 0; ii < Joysticks.Size(); ++ii)
 				{
-					if (Joysticks[i] == p)
+					if (Joysticks[ii] == p)
 					{
 						break;
 					}
 				}
-				if (i == Joysticks.Size())
+				if (ii == Joysticks.Size())
 				{
 					CurrentMenu->Close();
 				}

src/common/menu/menu.cpp

@@ -271,7 +271,7 @@ DMenu::DMenu(DMenu *parent)
 	DontDim = false;
 	GC::WriteBarrier(this, parent);
 }
-	
+
 //=============================================================================
 //
 //

src/common/menu/menudef.cpp

@@ -340,6 +340,10 @@ static void DoParseListMenuBody(FScanner &sc, DListMenuDescriptor *desc, bool &s
 		{
 			desc->mCenter = true;
 		}
+		else if (sc.Compare("Selecteditem"))
+		{
+			desc->mSelectedItem = desc->mItems.Size() - 1;
+		}
 		else if (sc.Compare("animatedtransition"))
 		{
 			desc->mAnimatedTransition = true;
@@ -563,8 +567,8 @@ static void DoParseListMenuBody(FScanner &sc, DListMenuDescriptor *desc, bool &s
 							// NB: index has been incremented, so we're not affecting the newly inserted item here.
 							for (unsigned int i = insertIndex; i < desc->mItems.Size(); i++)
 							{
-								auto item = desc->mItems[i];
+								auto litem = desc->mItems[i];
-								if (item->GetClass()->IsDescendantOf("ListMenuItemSelectable"))
+								if (litem->GetClass()->IsDescendantOf("ListMenuItemSelectable"))
 								{
 									desc->mItems[i]->mYpos += desc->mLinespacing;
 								}
@@ -660,9 +664,9 @@ static bool FindMatchingItem(DMenuItemBase *desc)
 	MenuDescriptorList::Pair *pair;
 	while (it.NextPair(pair))
 	{
-		for (auto it : pair->Value->mItems)
+		for (auto item : pair->Value->mItems)
 		{
-			if (it->mAction == name && GetGroup(it) == grp) return true;
+			if (item->mAction == name && GetGroup(item) == grp) return true;
 		}
 	}
 	return false;

src/common/menu/savegamemanager.cpp

@@ -151,8 +151,6 @@ int FSavegameManagerBase::InsertSaveNode(FSaveGameNode *node)
 
 void FSavegameManagerBase::NotifyNewSave(const FString &file, const FString &title, bool okForQuicksave, bool forceQuicksave)
 {
-	FSaveGameNode *node;
-
 	if (file.IsEmpty())
 		return;
 
@@ -180,7 +178,7 @@ void FSavegameManagerBase::NotifyNewSave(const FString &file, const FString &tit
 		}
 	}
 
-	node = new FSaveGameNode;
+	auto node = new FSaveGameNode;
 	node->SaveTitle = title;
 	node->Filename = file;
 	node->bOldVersion = false;