move x86 lzma loader patch from generic patches into rdc-2.6 - it's causing problems at least in qemu, maybe on other systems as well and generic x86 doesn't really need the extra compression

git-svn-id: svn://svn.openwrt.org/openwrt/trunk@7342 3c298f89-4303-0410-b956-a3cf2f4a3e73

1 files changed, 1035 insertions, 0 deletions

diff --git a/target/linux/rdc-2.6/patches/600-x86_lzma.patch b/target/linux/rdc-2.6/patches/600-x86_lzma.patch
new file mode 100644
index 0000000000..b370827b96
--- /dev/null
+++ b/target/linux/rdc-2.6/patches/600-x86_lzma.patch
@@ -0,0 +1,1035 @@
+diff -Naur linux-old/arch/i386/boot/compressed/LzmaDecode.c linux-lzma/arch/i386/boot/compressed/LzmaDecode.c
+--- linux-old/arch/i386/boot/compressed/LzmaDecode.c	1969-12-31 19:00:00.000000000 -0500
++++ linux-lzma/arch/i386/boot/compressed/LzmaDecode.c	2005-06-05 00:07:38.000000000 -0400
+@@ -0,0 +1,586 @@
++/*
++  LzmaDecode.c
++  LZMA Decoder (optimized for Speed version)
++  
++  LZMA SDK 4.17 Copyright (c) 1999-2005 Igor Pavlov (2005-04-05)
++  http://www.7-zip.org/
++
++  LZMA SDK is licensed under two licenses:
++  1) GNU Lesser General Public License (GNU LGPL)
++  2) Common Public License (CPL)
++  It means that you can select one of these two licenses and 
++  follow rules of that license.
++
++  SPECIAL EXCEPTION:
++  Igor Pavlov, as the author of this Code, expressly permits you to 
++  statically or dynamically link your Code (or bind by name) to the 
++  interfaces of this file without subjecting your linked Code to the 
++  terms of the CPL or GNU LGPL. Any modifications or additions 
++  to this file, however, are subject to the LGPL or CPL terms.
++*/
++
++#include "LzmaDecode.h"
++
++#ifndef Byte
++#define Byte unsigned char
++#endif
++
++#define kNumTopBits 24
++#define kTopValue ((UInt32)1 << kNumTopBits)
++
++#define kNumBitModelTotalBits 11
++#define kBitModelTotal (1 << kNumBitModelTotalBits)
++#define kNumMoveBits 5
++
++#define RC_READ_BYTE (*Buffer++)
++
++#define RC_INIT2 Code = 0; Range = 0xFFFFFFFF; \
++  { int i; for(i = 0; i < 5; i++) { RC_TEST; Code = (Code << 8) | RC_READ_BYTE; }}
++
++#ifdef _LZMA_IN_CB
++
++#define RC_TEST { if (Buffer == BufferLim) \
++  { UInt32 size; int result = InCallback->Read(InCallback, &Buffer, &size); if (result != LZMA_RESULT_OK) return result; \
++  BufferLim = Buffer + size; if (size == 0) return LZMA_RESULT_DATA_ERROR; }}
++
++#define RC_INIT Buffer = BufferLim = 0; RC_INIT2
++
++#else
++
++#define RC_TEST { if (Buffer == BufferLim) return LZMA_RESULT_DATA_ERROR; }
++
++#define RC_INIT(buffer, bufferSize) Buffer = buffer; BufferLim = buffer + bufferSize; RC_INIT2
++ 
++#endif
++
++#define RC_NORMALIZE if (Range < kTopValue) { RC_TEST; Range <<= 8; Code = (Code << 8) | RC_READ_BYTE; }
++
++#define IfBit0(p) RC_NORMALIZE; bound = (Range >> kNumBitModelTotalBits) * *(p); if (Code < bound)
++#define UpdateBit0(p) Range = bound; *(p) += (kBitModelTotal - *(p)) >> kNumMoveBits;
++#define UpdateBit1(p) Range -= bound; Code -= bound; *(p) -= (*(p)) >> kNumMoveBits;
++
++#define RC_GET_BIT2(p, mi, A0, A1) IfBit0(p) \
++  { UpdateBit0(p); mi <<= 1; A0; } else \
++  { UpdateBit1(p); mi = (mi + mi) + 1; A1; } 
++  
++#define RC_GET_BIT(p, mi) RC_GET_BIT2(p, mi, ; , ;)               
++
++#define RangeDecoderBitTreeDecode(probs, numLevels, res) \
++  { int i = numLevels; res = 1; \
++  do { CProb *p = probs + res; RC_GET_BIT(p, res) } while(--i != 0); \
++  res -= (1 << numLevels); }
++
++
++#define kNumPosBitsMax 4
++#define kNumPosStatesMax (1 << kNumPosBitsMax)
++
++#define kLenNumLowBits 3
++#define kLenNumLowSymbols (1 << kLenNumLowBits)
++#define kLenNumMidBits 3
++#define kLenNumMidSymbols (1 << kLenNumMidBits)
++#define kLenNumHighBits 8
++#define kLenNumHighSymbols (1 << kLenNumHighBits)
++
++#define LenChoice 0
++#define LenChoice2 (LenChoice + 1)
++#define LenLow (LenChoice2 + 1)
++#define LenMid (LenLow + (kNumPosStatesMax << kLenNumLowBits))
++#define LenHigh (LenMid + (kNumPosStatesMax << kLenNumMidBits))
++#define kNumLenProbs (LenHigh + kLenNumHighSymbols) 
++
++
++#define kNumStates 12
++#define kNumLitStates 7
++
++#define kStartPosModelIndex 4
++#define kEndPosModelIndex 14
++#define kNumFullDistances (1 << (kEndPosModelIndex >> 1))
++
++#define kNumPosSlotBits 6
++#define kNumLenToPosStates 4
++
++#define kNumAlignBits 4
++#define kAlignTableSize (1 << kNumAlignBits)
++
++#define kMatchMinLen 2
++
++#define IsMatch 0
++#define IsRep (IsMatch + (kNumStates << kNumPosBitsMax))
++#define IsRepG0 (IsRep + kNumStates)
++#define IsRepG1 (IsRepG0 + kNumStates)
++#define IsRepG2 (IsRepG1 + kNumStates)
++#define IsRep0Long (IsRepG2 + kNumStates)
++#define PosSlot (IsRep0Long + (kNumStates << kNumPosBitsMax))
++#define SpecPos (PosSlot + (kNumLenToPosStates << kNumPosSlotBits))
++#define Align (SpecPos + kNumFullDistances - kEndPosModelIndex)
++#define LenCoder (Align + kAlignTableSize)
++#define RepLenCoder (LenCoder + kNumLenProbs)
++#define Literal (RepLenCoder + kNumLenProbs)
++
++#if Literal != LZMA_BASE_SIZE
++StopCompilingDueBUG
++#endif
++
++#ifdef _LZMA_OUT_READ
++
++typedef struct _LzmaVarState
++{
++  Byte *Buffer;
++  Byte *BufferLim;
++  UInt32 Range;
++  UInt32 Code;
++  #ifdef _LZMA_IN_CB
++  ILzmaInCallback *InCallback;
++  #endif
++  Byte *Dictionary;
++  UInt32 DictionarySize;
++  UInt32 DictionaryPos;
++  UInt32 GlobalPos;
++  UInt32 Reps[4];
++  int lc;
++  int lp;
++  int pb;
++  int State;
++  int RemainLen;
++  Byte TempDictionary[4];
++} LzmaVarState;
++
++int LzmaDecoderInit(
++    unsigned char *buffer, UInt32 bufferSize,
++    int lc, int lp, int pb,
++    unsigned char *dictionary, UInt32 dictionarySize,
++    #ifdef _LZMA_IN_CB
++    ILzmaInCallback *InCallback
++    #else
++    unsigned char *inStream, UInt32 inSize
++    #endif
++    )
++{
++  Byte *Buffer;
++  Byte *BufferLim;
++  UInt32 Range;
++  UInt32 Code;
++  LzmaVarState *vs = (LzmaVarState *)buffer;
++  CProb *p = (CProb *)(buffer + sizeof(LzmaVarState));
++  UInt32 numProbs = Literal + ((UInt32)LZMA_LIT_SIZE << (lc + lp));
++  UInt32 i;
++  if (bufferSize < numProbs * sizeof(CProb) + sizeof(LzmaVarState))
++    return LZMA_RESULT_NOT_ENOUGH_MEM;
++  vs->Dictionary = dictionary;
++  vs->DictionarySize = dictionarySize;
++  vs->DictionaryPos = 0;
++  vs->GlobalPos = 0;
++  vs->Reps[0] = vs->Reps[1] = vs->Reps[2] = vs->Reps[3] = 1;
++  vs->lc = lc;
++  vs->lp = lp;
++  vs->pb = pb;
++  vs->State = 0;
++  vs->RemainLen = 0;
++  dictionary[dictionarySize - 1] = 0;
++  for (i = 0; i < numProbs; i++)
++    p[i] = kBitModelTotal >> 1; 
++
++  #ifdef _LZMA_IN_CB
++  RC_INIT;
++  #else
++  RC_INIT(inStream, inSize);
++  #endif
++  vs->Buffer = Buffer;
++  vs->BufferLim = BufferLim;
++  vs->Range = Range;
++  vs->Code = Code;
++  #ifdef _LZMA_IN_CB
++  vs->InCallback = InCallback;
++  #endif
++
++  return LZMA_RESULT_OK;
++}
++
++int LzmaDecode(unsigned char *buffer, 
++    unsigned char *outStream, UInt32 outSize,
++    UInt32 *outSizeProcessed)
++{
++  LzmaVarState *vs = (LzmaVarState *)buffer;
++  Byte *Buffer = vs->Buffer;
++  Byte *BufferLim = vs->BufferLim;
++  UInt32 Range = vs->Range;
++  UInt32 Code = vs->Code;
++  #ifdef _LZMA_IN_CB
++  ILzmaInCallback *InCallback = vs->InCallback;
++  #endif
++  CProb *p = (CProb *)(buffer + sizeof(LzmaVarState));
++  int state = vs->State;
++  Byte previousByte;
++  UInt32 rep0 = vs->Reps[0], rep1 = vs->Reps[1], rep2 = vs->Reps[2], rep3 = vs->Reps[3];
++  UInt32 nowPos = 0;
++  UInt32 posStateMask = (1 << (vs->pb)) - 1;
++  UInt32 literalPosMask = (1 << (vs->lp)) - 1;
++  int lc = vs->lc;
++  int len = vs->RemainLen;
++  UInt32 globalPos = vs->GlobalPos;
++
++  Byte *dictionary = vs->Dictionary;
++  UInt32 dictionarySize = vs->DictionarySize;
++  UInt32 dictionaryPos = vs->DictionaryPos;
++
++  Byte tempDictionary[4];
++  if (dictionarySize == 0)
++  {
++    dictionary = tempDictionary;
++    dictionarySize = 1;
++    tempDictionary[0] = vs->TempDictionary[0];
++  }
++
++  if (len == -1)
++  {
++    *outSizeProcessed = 0;
++    return LZMA_RESULT_OK;
++  }
++
++  while(len != 0 && nowPos < outSize)
++  {
++    UInt32 pos = dictionaryPos - rep0;
++    if (pos >= dictionarySize)
++      pos += dictionarySize;
++    outStream[nowPos++] = dictionary[dictionaryPos] = dictionary[pos];
++    if (++dictionaryPos == dictionarySize)
++      dictionaryPos = 0;
++    len--;
++  }
++  if (dictionaryPos == 0)
++    previousByte = dictionary[dictionarySize - 1];
++  else
++    previousByte = dictionary[dictionaryPos - 1];
++#else
++
++int LzmaDecode(
++    Byte *buffer, UInt32 bufferSize,
++    int lc, int lp, int pb,
++    #ifdef _LZMA_IN_CB
++    ILzmaInCallback *InCallback,
++    #else
++    unsigned char *inStream, UInt32 inSize,
++    #endif
++    unsigned char *outStream, UInt32 outSize,
++    UInt32 *outSizeProcessed)
++{
++  UInt32 numProbs = Literal + ((UInt32)LZMA_LIT_SIZE << (lc + lp));
++  CProb *p = (CProb *)buffer;
++
++  UInt32 i;
++  int state = 0;
++  Byte previousByte = 0;
++  UInt32 rep0 = 1, rep1 = 1, rep2 = 1, rep3 = 1;
++  UInt32 nowPos = 0;
++  UInt32 posStateMask = (1 << pb) - 1;
++  UInt32 literalPosMask = (1 << lp) - 1;
++  int len = 0;
++  
++  Byte *Buffer;
++  Byte *BufferLim;
++  UInt32 Range;
++  UInt32 Code;
++  
++  if (bufferSize < numProbs * sizeof(CProb))
++    return LZMA_RESULT_NOT_ENOUGH_MEM;
++  for (i = 0; i < numProbs; i++)
++    p[i] = kBitModelTotal >> 1;
++  
++
++  #ifdef _LZMA_IN_CB
++  RC_INIT;
++  #else
++  RC_INIT(inStream, inSize);
++  #endif
++#endif
++
++  *outSizeProcessed = 0;
++  while(nowPos < outSize)
++  {
++    CProb *prob;
++    UInt32 bound;
++    int posState = (int)(
++        (nowPos 
++        #ifdef _LZMA_OUT_READ
++        + globalPos
++        #endif
++        )
++        & posStateMask);
++
++    prob = p + IsMatch + (state << kNumPosBitsMax) + posState;
++    IfBit0(prob)
++    {
++      int symbol = 1;
++      UpdateBit0(prob)
++      prob = p + Literal + (LZMA_LIT_SIZE * 
++        (((
++        (nowPos 
++        #ifdef _LZMA_OUT_READ
++        + globalPos
++        #endif
++        )
++        & literalPosMask) << lc) + (previousByte >> (8 - lc))));
++
++      if (state >= kNumLitStates)
++      {
++        int matchByte;
++        #ifdef _LZMA_OUT_READ
++        UInt32 pos = dictionaryPos - rep0;
++        if (pos >= dictionarySize)
++          pos += dictionarySize;
++        matchByte = dictionary[pos];
++        #else
++        matchByte = outStream[nowPos - rep0];
++        #endif
++        do
++        {
++          int bit;
++          CProb *probLit;
++          matchByte <<= 1;
++          bit = (matchByte & 0x100);
++          probLit = prob + 0x100 + bit + symbol;
++          RC_GET_BIT2(probLit, symbol, if (bit != 0) break, if (bit == 0) break)
++        }
++        while (symbol < 0x100);
++      }
++      while (symbol < 0x100)
++      {
++        CProb *probLit = prob + symbol;
++        RC_GET_BIT(probLit, symbol)
++      }
++      previousByte = (Byte)symbol;
++
++      outStream[nowPos++] = previousByte;
++      #ifdef _LZMA_OUT_READ
++      dictionary[dictionaryPos] = previousByte;
++      if (++dictionaryPos == dictionarySize)
++        dictionaryPos = 0;
++      #endif
++      if (state < 4) state = 0;
++      else if (state < 10) state -= 3;
++      else state -= 6;
++    }
++    else             
++    {
++      UpdateBit1(prob);
++      prob = p + IsRep + state;
++      IfBit0(prob)
++      {
++        UpdateBit0(prob);
++        rep3 = rep2;
++        rep2 = rep1;
++        rep1 = rep0;
++        state = state < kNumLitStates ? 0 : 3;
++        prob = p + LenCoder;
++      }
++      else
++      {
++        UpdateBit1(prob);
++        prob = p + IsRepG0 + state;
++        IfBit0(prob)
++        {
++          UpdateBit0(prob);
++          prob = p + IsRep0Long + (state << kNumPosBitsMax) + posState;
++          IfBit0(prob)
++          {
++            #ifdef _LZMA_OUT_READ
++            UInt32 pos;
++            #endif
++            UpdateBit0(prob);
++            if (nowPos 
++                #ifdef _LZMA_OUT_READ
++                + globalPos
++                #endif
++                == 0)
++              return LZMA_RESULT_DATA_ERROR;
++            state = state < kNumLitStates ? 9 : 11;
++            #ifdef _LZMA_OUT_READ
++            pos = dictionaryPos - rep0;
++            if (pos >= dictionarySize)
++              pos += dictionarySize;
++            previousByte = dictionary[pos];
++            dictionary[dictionaryPos] = previousByte;
++            if (++dictionaryPos == dictionarySize)
++              dictionaryPos = 0;
++            #else
++            previousByte = outStream[nowPos - rep0];
++            #endif
++            outStream[nowPos++] = previousByte;
++            continue;
++          }
++          else
++          {
++            UpdateBit1(prob);
++          }
++        }
++        else
++        {
++          UInt32 distance;
++          UpdateBit1(prob);
++          prob = p + IsRepG1 + state;
++          IfBit0(prob)
++          {
++            UpdateBit0(prob);
++            distance = rep1;
++          }
++          else 
++          {
++            UpdateBit1(prob);
++            prob = p + IsRepG2 + state;
++            IfBit0(prob)
++            {
++              UpdateBit0(prob);
++              distance = rep2;
++            }
++            else
++            {
++              UpdateBit1(prob);
++              distance = rep3;
++              rep3 = rep2;
++            }
++            rep2 = rep1;
++          }
++          rep1 = rep0;
++          rep0 = distance;
++        }
++        state = state < kNumLitStates ? 8 : 11;
++        prob = p + RepLenCoder;
++      }
++      {
++        int numBits, offset;
++        CProb *probLen = prob + LenChoice;
++        IfBit0(probLen)
++        {
++          UpdateBit0(probLen);
++          probLen = prob + LenLow + (posState << kLenNumLowBits);
++          offset = 0;
++          numBits = kLenNumLowBits;
++        }
++        else
++        {
++          UpdateBit1(probLen);
++          probLen = prob + LenChoice2;
++          IfBit0(probLen)
++          {
++            UpdateBit0(probLen);
++            probLen = prob + LenMid + (posState << kLenNumMidBits);
++            offset = kLenNumLowSymbols;
++            numBits = kLenNumMidBits;
++          }
++          else
++          {
++            UpdateBit1(probLen);
++            probLen = prob + LenHigh;
++            offset = kLenNumLowSymbols + kLenNumMidSymbols;
++            numBits = kLenNumHighBits;
++          }
++        }
++        RangeDecoderBitTreeDecode(probLen, numBits, len);
++        len += offset;
++      }
++
++      if (state < 4)
++      {
++        int posSlot;
++        state += kNumLitStates;
++        prob = p + PosSlot +
++            ((len < kNumLenToPosStates ? len : kNumLenToPosStates - 1) << 
++            kNumPosSlotBits);
++        RangeDecoderBitTreeDecode(prob, kNumPosSlotBits, posSlot);
++        if (posSlot >= kStartPosModelIndex)
++        {
++          int numDirectBits = ((posSlot >> 1) - 1);
++          rep0 = (2 | ((UInt32)posSlot & 1));
++          if (posSlot < kEndPosModelIndex)
++          {
++            rep0 <<= numDirectBits;
++            prob = p + SpecPos + rep0 - posSlot - 1;
++          }
++          else
++          {
++            numDirectBits -= kNumAlignBits;
++            do
++            {
++              RC_NORMALIZE
++              Range >>= 1;
++              rep0 <<= 1;
++              if (Code >= Range)
++              {
++                Code -= Range;
++                rep0 |= 1;
++              }
++            }
++            while (--numDirectBits != 0);
++            prob = p + Align;
++            rep0 <<= kNumAlignBits;
++            numDirectBits = kNumAlignBits;
++          }
++          {
++            int i = 1;
++            int mi = 1;
++            do
++            {
++              CProb *prob3 = prob + mi;
++              RC_GET_BIT2(prob3, mi, ; , rep0 |= i);
++              i <<= 1;
++            }
++            while(--numDirectBits != 0);
++          }
++        }
++        else
++          rep0 = posSlot;
++        if (++rep0 == (UInt32)(0))
++        {
++          /* it's for stream version */
++          len = -1;
++          break;
++        }
++      }
++
++      len += kMatchMinLen;
++      if (rep0 > nowPos 
++        #ifdef _LZMA_OUT_READ
++        + globalPos || rep0 > dictionarySize
++        #endif
++        ) 
++        return LZMA_RESULT_DATA_ERROR;
++      do
++      {
++        #ifdef _LZMA_OUT_READ
++        UInt32 pos = dictionaryPos - rep0;
++        if (pos >= dictionarySize)
++          pos += dictionarySize;
++        previousByte = dictionary[pos];
++        dictionary[dictionaryPos] = previousByte;
++        if (++dictionaryPos == dictionarySize)
++          dictionaryPos = 0;
++        #else
++        previousByte = outStream[nowPos - rep0];
++        #endif
++        len--;
++        outStream[nowPos++] = previousByte;
++      }
++      while(len != 0 && nowPos < outSize);
++    }
++  }
++  RC_NORMALIZE;
++
++  #ifdef _LZMA_OUT_READ
++  vs->Buffer = Buffer;
++  vs->BufferLim = BufferLim;
++  vs->Range = Range;
++  vs->Code = Code;
++  vs->DictionaryPos = dictionaryPos;
++  vs->GlobalPos = globalPos + nowPos;
++  vs->Reps[0] = rep0;
++  vs->Reps[1] = rep1;
++  vs->Reps[2] = rep2;
++  vs->Reps[3] = rep3;
++  vs->State = state;
++  vs->RemainLen = len;
++  vs->TempDictionary[0] = tempDictionary[0];
++  #endif
++
++  *outSizeProcessed = nowPos;
++  return LZMA_RESULT_OK;
++}
+diff -Naur linux-old/arch/i386/boot/compressed/LzmaDecode.h linux-lzma/arch/i386/boot/compressed/LzmaDecode.h
+--- linux-old/arch/i386/boot/compressed/LzmaDecode.h	1969-12-31 19:00:00.000000000 -0500
++++ linux-lzma/arch/i386/boot/compressed/LzmaDecode.h	2005-06-05 00:07:39.000000000 -0400
+@@ -0,0 +1,100 @@
++/* 
++  LzmaDecode.h
++  LZMA Decoder interface
++
++  LZMA SDK 4.16 Copyright (c) 1999-2005 Igor Pavlov (2005-03-18)
++  http://www.7-zip.org/
++
++  LZMA SDK is licensed under two licenses:
++  1) GNU Lesser General Public License (GNU LGPL)
++  2) Common Public License (CPL)
++  It means that you can select one of these two licenses and 
++  follow rules of that license.
++
++  SPECIAL EXCEPTION:
++  Igor Pavlov, as the author of this code, expressly permits you to 
++  statically or dynamically link your code (or bind by name) to the 
++  interfaces of this file without subjecting your linked code to the 
++  terms of the CPL or GNU LGPL. Any modifications or additions 
++  to this file, however, are subject to the LGPL or CPL terms.
++*/
++
++#ifndef __LZMADECODE_H
++#define __LZMADECODE_H
++
++/* #define _LZMA_IN_CB */
++/* Use callback for input data */
++
++/* #define _LZMA_OUT_READ */
++/* Use read function for output data */
++
++/* #define _LZMA_PROB32 */
++/* It can increase speed on some 32-bit CPUs, 
++   but memory usage will be doubled in that case */
++
++/* #define _LZMA_LOC_OPT */
++/* Enable local speed optimizations inside code */
++
++#ifndef UInt32
++#ifdef _LZMA_UINT32_IS_ULONG
++#define UInt32 unsigned long
++#else
++#define UInt32 unsigned int
++#endif
++#endif
++
++#ifdef _LZMA_PROB32
++#define CProb UInt32
++#else
++#define CProb unsigned short
++#endif
++
++#define LZMA_RESULT_OK 0
++#define LZMA_RESULT_DATA_ERROR 1
++#define LZMA_RESULT_NOT_ENOUGH_MEM 2
++
++#ifdef _LZMA_IN_CB
++typedef struct _ILzmaInCallback
++{
++  int (*Read)(void *object, unsigned char **buffer, UInt32 *bufferSize);
++} ILzmaInCallback;
++#endif
++
++#define LZMA_BASE_SIZE 1846
++#define LZMA_LIT_SIZE 768
++
++/* 
++bufferSize = (LZMA_BASE_SIZE + (LZMA_LIT_SIZE << (lc + lp)))* sizeof(CProb)
++bufferSize += 100 in case of _LZMA_OUT_READ
++by default CProb is unsigned short, 
++but if specify _LZMA_PROB_32, CProb will be UInt32(unsigned int)
++*/
++
++#ifdef _LZMA_OUT_READ
++int LzmaDecoderInit(
++    unsigned char *buffer, UInt32 bufferSize,
++    int lc, int lp, int pb,
++    unsigned char *dictionary, UInt32 dictionarySize,
++  #ifdef _LZMA_IN_CB
++    ILzmaInCallback *inCallback
++  #else
++    unsigned char *inStream, UInt32 inSize
++  #endif
++);
++#endif
++
++int LzmaDecode(
++    unsigned char *buffer, 
++  #ifndef _LZMA_OUT_READ
++    UInt32 bufferSize,
++    int lc, int lp, int pb,
++  #ifdef _LZMA_IN_CB
++    ILzmaInCallback *inCallback,
++  #else
++    unsigned char *inStream, UInt32 inSize,
++  #endif
++  #endif
++    unsigned char *outStream, UInt32 outSize,
++    UInt32 *outSizeProcessed);
++
++#endif
+diff -Naur linux-old/arch/i386/boot/compressed/lzma_misc.c linux-lzma/arch/i386/boot/compressed/lzma_misc.c
+--- linux-old/arch/i386/boot/compressed/lzma_misc.c	1969-12-31 19:00:00.000000000 -0500
++++ linux-lzma/arch/i386/boot/compressed/lzma_misc.c	2005-06-04 21:33:48.000000000 -0400
+@@ -0,0 +1,281 @@
++/*
++ * lzma_misc.c
++ * 
++ * malloc by Hannu Savolainen 1993 and Matthias Urlichs 1994
++ * puts by Nick Holloway 1993, better puts by Martin Mares 1995
++ * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996
++ * 
++ * Decompress LZMA compressed vmlinuz 
++ * Version 0.9 Copyright (c) Ming-Ching Tiew mctiew@yahoo.com
++ * Program adapted from misc.c for 2.6 kernel
++ * Forward ported to latest 2.6 version of misc.c by
++ * Felix Fietkau <nbd@openwrt.org>
++ */
++
++#undef CONFIG_PARAVIRT
++#include <linux/linkage.h>
++#include <linux/vmalloc.h>
++#include <linux/screen_info.h>
++#include <asm/io.h>
++#include <asm/page.h>
++#include <asm/boot.h>
++
++/* WARNING!!
++ * This code is compiled with -fPIC and it is relocated dynamically
++ * at run time, but no relocation processing is performed.
++ * This means that it is not safe to place pointers in static structures.
++ */
++
++/*
++ * Getting to provable safe in place decompression is hard.
++ * Worst case behaviours need to be analized.
++ * Background information:
++ *
++ * The file layout is:
++ *    magic[2]
++ *    method[1]
++ *    flags[1]
++ *    timestamp[4]
++ *    extraflags[1]
++ *    os[1]
++ *    compressed data blocks[N]
++ *    crc[4] orig_len[4]
++ *
++ * resulting in 18 bytes of non compressed data overhead.
++ *
++ * Files divided into blocks
++ * 1 bit (last block flag)
++ * 2 bits (block type)
++ *
++ * 1 block occurs every 32K -1 bytes or when there 50% compression has been achieved.
++ * The smallest block type encoding is always used.
++ *
++ * stored:
++ *    32 bits length in bytes.
++ *
++ * fixed:
++ *    magic fixed tree.
++ *    symbols.
++ *
++ * dynamic:
++ *    dynamic tree encoding.
++ *    symbols.
++ *
++ *
++ * The buffer for decompression in place is the length of the
++ * uncompressed data, plus a small amount extra to keep the algorithm safe.
++ * The compressed data is placed at the end of the buffer.  The output
++ * pointer is placed at the start of the buffer and the input pointer
++ * is placed where the compressed data starts.  Problems will occur
++ * when the output pointer overruns the input pointer.
++ *
++ * The output pointer can only overrun the input pointer if the input
++ * pointer is moving faster than the output pointer.  A condition only
++ * triggered by data whose compressed form is larger than the uncompressed
++ * form.
++ *
++ * The worst case at the block level is a growth of the compressed data
++ * of 5 bytes per 32767 bytes.
++ *
++ * The worst case internal to a compressed block is very hard to figure.
++ * The worst case can at least be boundined by having one bit that represents
++ * 32764 bytes and then all of the rest of the bytes representing the very
++ * very last byte.
++ *
++ * All of which is enough to compute an amount of extra data that is required
++ * to be safe.  To avoid problems at the block level allocating 5 extra bytes
++ * per 32767 bytes of data is sufficient.  To avoind problems internal to a block
++ * adding an extra 32767 bytes (the worst case uncompressed block size) is
++ * sufficient, to ensure that in the worst case the decompressed data for
++ * block will stop the byte before the compressed data for a block begins.
++ * To avoid problems with the compressed data's meta information an extra 18
++ * bytes are needed.  Leading to the formula:
++ *
++ * extra_bytes = (uncompressed_size >> 12) + 32768 + 18 + decompressor_size.
++ *
++ * Adding 8 bytes per 32K is a bit excessive but much easier to calculate.
++ * Adding 32768 instead of 32767 just makes for round numbers.
++ * Adding the decompressor_size is necessary as it musht live after all
++ * of the data as well.  Last I measured the decompressor is about 14K.
++ * 10K of actuall data and 4K of bss.
++ *
++ */
++
++/*
++ * gzip declarations
++ */
++
++#define OF(args)  args
++#define STATIC static
++
++#undef memcpy
++
++typedef unsigned char  uch;
++typedef unsigned short ush;
++typedef unsigned long  ulg;
++
++#define WSIZE 0x80000000	/* Window size must be at least 32k,
++				 * and a power of two
++				 * We don't actually have a window just
++				 * a huge output buffer so I report
++				 * a 2G windows size, as that should
++				 * always be larger than our output buffer.
++				 */
++
++static uch *inbuf;	/* input buffer */
++static uch *window;	/* Sliding window buffer, (and final output buffer) */
++
++static unsigned insize;  /* valid bytes in inbuf */
++static unsigned inptr;   /* index of next byte to be processed in inbuf */
++
++#define get_byte()  (inptr < insize ? inbuf[inptr++] : fill_inbuf())
++		
++/* Diagnostic functions */
++#ifdef DEBUG
++#  define Assert(cond,msg) {if(!(cond)) error(msg);}
++#  define Trace(x) fprintf x
++#  define Tracev(x) {if (verbose) fprintf x ;}
++#  define Tracevv(x) {if (verbose>1) fprintf x ;}
++#  define Tracec(c,x) {if (verbose && (c)) fprintf x ;}
++#  define Tracecv(c,x) {if (verbose>1 && (c)) fprintf x ;}
++#else
++#  define Assert(cond,msg)
++#  define Trace(x)
++#  define Tracev(x)
++#  define Tracevv(x)
++#  define Tracec(c,x)
++#  define Tracecv(c,x)
++#endif
++
++static int  fill_inbuf(void);
++  
++/*
++ * This is set up by the setup-routine at boot-time
++ */
++static unsigned char *real_mode; /* Pointer to real-mode data */
++extern unsigned char input_data[];
++extern int input_len;
++
++static void error(char *x);
++static void *memcpy(void *dest, const void *src, unsigned n);
++
++#ifdef CONFIG_X86_NUMAQ
++void *xquad_portio;
++#endif
++
++static void* memcpy(void* dest, const void* src, unsigned n)
++{
++	int i;
++	char *d = (char *)dest, *s = (char *)src;
++
++	for (i=0;i<n;i++) d[i] = s[i];
++	return dest;
++}
++
++/* ===========================================================================
++ * Fill the input buffer. This is called only when the buffer is empty
++ * and at least one byte is really needed.
++ */
++static int fill_inbuf(void)
++{
++	error("ran out of input data");
++	return 0;
++}
++
++
++// When using LZMA in callback, the compressed length is not needed.
++// Otherwise you need a special version of lzma compression program
++// which will pad the compressed length in the header.
++#define _LZMA_IN_CB
++#include "LzmaDecode.h"
++#include "LzmaDecode.c"
++
++static int read_byte(void *object, unsigned char **buffer, UInt32 *bufferSize);
++
++
++/*
++ * Do the lzma decompression
++ * When using LZMA in callback, the end of input stream is automatically determined
++ */
++static int lzma_unzip(void)
++{
++
++	unsigned int i;  /* temp value */
++	unsigned int lc; /* literal context bits */
++	unsigned int lp; /* literal pos state bits */
++	unsigned int pb; /* pos state bits */
++	unsigned char* workspace;
++	unsigned int uncompressedSize = 0;
++	unsigned char* p;
++	
++	ILzmaInCallback callback;
++	callback.Read = read_byte;
++
++	/* lzma args */
++	i = get_byte();
++	lc = i % 9, i = i / 9;
++	lp = i % 5, pb = i / 5;
++	
++	/* skip dictionary size */
++	for (i = 0; i < 4; i++) 
++		get_byte();
++	// get uncompressedSize 	
++	p= (char*)&uncompressedSize;	
++	for (i = 0; i < 4; i++) 
++	    *p++ = get_byte();
++	    
++	//get compressedSize 
++	for (i = 0; i < 4; i++) 
++		get_byte();
++	
++	// point it beyond uncompresedSize
++	workspace = window + uncompressedSize;
++	
++	/* decompress kernel */
++	if (LzmaDecode(workspace, ~0, lc, lp, pb, &callback,
++		(unsigned char*)window, uncompressedSize, &i) == LZMA_RESULT_OK)
++		return 0;
++	else
++		return 1;
++}
++
++
++#ifdef  _LZMA_IN_CB
++static int read_byte(void *object, unsigned char **buffer, UInt32 *bufferSize)
++{
++	static unsigned int i = 0;
++	static unsigned char val;
++	*bufferSize = 1;
++	val = get_byte();
++	*buffer = &val;
++	return LZMA_RESULT_OK;
++}	
++#endif
++
++static void error(char *x)
++{
++	while(1);	/* Halt */
++}
++
++asmlinkage void decompress_kernel(void *rmode, unsigned long end,
++			uch *input_data, unsigned long input_len, uch *output)
++{
++	real_mode = rmode;
++
++	window = output;
++	inbuf  = input_data;	/* Input buffer */
++	insize = input_len;
++	inptr  = 0;
++
++	if ((u32)output & (CONFIG_PHYSICAL_ALIGN -1))
++		error("Destination address not CONFIG_PHYSICAL_ALIGN aligned");
++	if (end > ((-__PAGE_OFFSET-(512 <<20)-1) & 0x7fffffff))
++		error("Destination address too large");
++#ifndef CONFIG_RELOCATABLE
++	if ((u32)output != LOAD_PHYSICAL_ADDR)
++		error("Wrong destination address");
++#endif
++
++	lzma_unzip();
++	return;
++}
+diff -Naur linux-old/arch/i386/boot/compressed/Makefile linux-lzma/arch/i386/boot/compressed/Makefile
+--- linux-old/arch/i386/boot/compressed/Makefile	2005-06-04 21:53:40.000000000 -0400
++++ linux-lzma/arch/i386/boot/compressed/Makefile	2005-06-05 00:25:23.000000000 -0400
+@@ -4,15 +4,15 @@
+ # create a compressed vmlinux image from the original vmlinux
+ #
+ 
+-targets		:= vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o \
++targets		:= vmlinux vmlinux.bin vmlinux.bin.lzma head.o lzma_misc.o piggy.o \
+ 			vmlinux.bin.all vmlinux.relocs
+ EXTRA_AFLAGS	:= -traditional
+ 
+ LDFLAGS_vmlinux := -T
+-CFLAGS_misc.o += -fPIC
++CFLAGS_lzma_misc.o += -fPIC
+ hostprogs-y	:= relocs
+ 
+-$(obj)/vmlinux: $(src)/vmlinux.lds $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o FORCE
++$(obj)/vmlinux: $(src)/vmlinux.lds $(obj)/head.o $(obj)/lzma_misc.o $(obj)/piggy.o FORCE
+ 	$(call if_changed,ld)
+ 	@:
+ 
+@@ -32,14 +32,14 @@
+ 	$(call if_changed,relocbin)
+ 
+ ifdef CONFIG_RELOCATABLE
+-$(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin.all FORCE
+-	$(call if_changed,gzip)
++$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin.all FORCE
++	$(call if_changed,lzma)
+ else
+-$(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE
+-	$(call if_changed,gzip)
++$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin FORCE
++	$(call if_changed,lzma)
+ endif
+ 
+ LDFLAGS_piggy.o := -r --format binary --oformat elf32-i386 -T
+ 
+-$(obj)/piggy.o: $(src)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE
++$(obj)/piggy.o: $(src)/vmlinux.scr $(obj)/vmlinux.bin.lzma FORCE
+ 	$(call if_changed,ld)
+diff -urN linux-2.6.19.2/scripts/Makefile.lib linux-2.6.19.2.new/scripts/Makefile.lib
+--- linux-2.6.19.2/scripts/Makefile.lib	2007-01-10 20:10:37.000000000 +0100
++++ linux-2.6.19.2.new/scripts/Makefile.lib	2007-04-15 23:51:54.000000000 +0200
+@@ -162,4 +162,9 @@
+ quiet_cmd_gzip = GZIP    $@
+ cmd_gzip = gzip -f -9 < $< > $@
+ 
+-
++# LZMA
++#
++quiet_cmd_lzma = LZMA $@
++cmd_lzma = lzma e $< $@ -lc7 -lp0 -pb0
++# to use lzmacomp,
++# cmd_lzma = lzmacomp $< 700 > $@