extlz4 0.3 → 0.3.1

data/contrib/lz4/lib/lz4.h

@@ -46,7 +46,7 @@ extern "C" {
 /**
   Introduction
 
-  LZ4 is lossless compression algorithm, providing compression speed at 500 MB/s per core,
+  LZ4 is lossless compression algorithm, providing compression speed >500 MB/s per core,
   scalable with multi-cores CPU. It features an extremely fast decoder, with speed in
   multiple GB/s per core, typically reaching RAM speed limits on multi-core systems.
 
@@ -58,16 +58,19 @@ extern "C" {
     - unbounded multiple steps (described as Streaming compression)
 
   lz4.h generates and decodes LZ4-compressed blocks (doc/lz4_Block_format.md).
-  Decompressing a block requires additional metadata, such as its compressed size.
+  Decompressing such a compressed block requires additional metadata.
+  Exact metadata depends on exact decompression function.
+  For the typical case of LZ4_decompress_safe(),
+  metadata includes block's compressed size, and maximum bound of decompressed size.
   Each application is free to encode and pass such metadata in whichever way it wants.
 
   lz4.h only handle blocks, it can not generate Frames.
 
   Blocks are different from Frames (doc/lz4_Frame_format.md).
   Frames bundle both blocks and metadata in a specified manner.
-  This are required for compressed data to be self-contained and portable.
+  Embedding metadata is required for compressed data to be self-contained and portable.
   Frame format is delivered through a companion API, declared in lz4frame.h.
-  Note that the `lz4` CLI can only manage frames.
+  The `lz4` CLI can only manage frames.
 */
 
 /*^***************************************************************
@@ -97,7 +100,7 @@ extern "C" {
 /*------   Version   ------*/
 #define LZ4_VERSION_MAJOR    1    /* for breaking interface changes  */
 #define LZ4_VERSION_MINOR    9    /* for new (non-breaking) interface capabilities */
-#define LZ4_VERSION_RELEASE  0    /* for tweaks, bug-fixes, or development */
+#define LZ4_VERSION_RELEASE  2    /* for tweaks, bug-fixes, or development */
 
 #define LZ4_VERSION_NUMBER (LZ4_VERSION_MAJOR *100*100 + LZ4_VERSION_MINOR *100 + LZ4_VERSION_RELEASE)
 
@@ -129,29 +132,35 @@ LZ4LIB_API const char* LZ4_versionString (void);   /**< library version string;
 *  Simple Functions
 **************************************/
 /*! LZ4_compress_default() :
-    Compresses 'srcSize' bytes from buffer 'src'
-    into already allocated 'dst' buffer of size 'dstCapacity'.
-    Compression is guaranteed to succeed if 'dstCapacity' >= LZ4_compressBound(srcSize).
-    It also runs faster, so it's a recommended setting.
-    If the function cannot compress 'src' into a more limited 'dst' budget,
-    compression stops *immediately*, and the function result is zero.
-    In which case, 'dst' content is undefined (invalid).
-        srcSize : max supported value is LZ4_MAX_INPUT_SIZE.
-        dstCapacity : size of buffer 'dst' (which must be already allocated)
-       @return  : the number of bytes written into buffer 'dst' (necessarily <= dstCapacity)
-                  or 0 if compression fails
-    Note : This function is protected against buffer overflow scenarios (never writes outside 'dst' buffer, nor read outside 'source' buffer).
-*/
+ *  Compresses 'srcSize' bytes from buffer 'src'
+ *  into already allocated 'dst' buffer of size 'dstCapacity'.
+ *  Compression is guaranteed to succeed if 'dstCapacity' >= LZ4_compressBound(srcSize).
+ *  It also runs faster, so it's a recommended setting.
+ *  If the function cannot compress 'src' into a more limited 'dst' budget,
+ *  compression stops *immediately*, and the function result is zero.
+ *  In which case, 'dst' content is undefined (invalid).
+ *      srcSize : max supported value is LZ4_MAX_INPUT_SIZE.
+ *      dstCapacity : size of buffer 'dst' (which must be already allocated)
+ *     @return  : the number of bytes written into buffer 'dst' (necessarily <= dstCapacity)
+ *                or 0 if compression fails
+ * Note : This function is protected against buffer overflow scenarios (never writes outside 'dst' buffer, nor read outside 'source' buffer).
+ */
 LZ4LIB_API int LZ4_compress_default(const char* src, char* dst, int srcSize, int dstCapacity);
 
 /*! LZ4_decompress_safe() :
-    compressedSize : is the exact complete size of the compressed block.
-    dstCapacity : is the size of destination buffer, which must be already allocated.
-   @return : the number of bytes decompressed into destination buffer (necessarily <= dstCapacity)
-             If destination buffer is not large enough, decoding will stop and output an error code (negative value).
-             If the source stream is detected malformed, the function will stop decoding and return a negative result.
-    Note : This function is protected against malicious data packets (never writes outside 'dst' buffer, nor read outside 'source' buffer).
-*/
+ *  compressedSize : is the exact complete size of the compressed block.
+ *  dstCapacity : is the size of destination buffer (which must be already allocated), presumed an upper bound of decompressed size.
+ * @return : the number of bytes decompressed into destination buffer (necessarily <= dstCapacity)
+ *           If destination buffer is not large enough, decoding will stop and output an error code (negative value).
+ *           If the source stream is detected malformed, the function will stop decoding and return a negative result.
+ * Note 1 : This function is protected against malicious data packets :
+ *          it will never writes outside 'dst' buffer, nor read outside 'source' buffer,
+ *          even if the compressed block is maliciously modified to order the decoder to do these actions.
+ *          In such case, the decoder stops immediately, and considers the compressed block malformed.
+ * Note 2 : compressedSize and dstCapacity must be provided to the function, the compressed block does not contain them.
+ *          The implementation is free to send / store / derive this information in whichever way is most beneficial.
+ *          If there is a need for a different format which bundles together both compressed data and its metadata, consider looking at lz4frame.h instead.
+ */
 LZ4LIB_API int LZ4_decompress_safe (const char* src, char* dst, int compressedSize, int dstCapacity);
 
 
@@ -388,6 +397,8 @@ LZ4LIB_API int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecod
  */
 LZ4LIB_API int LZ4_decompress_safe_usingDict (const char* src, char* dst, int srcSize, int dstCapcity, const char* dictStart, int dictSize);
 
+#endif /* LZ4_H_2983827168210 */
+
 
 /*^*************************************
  * !!!!!!   STATIC LINKING ONLY   !!!!!!
@@ -413,14 +424,17 @@ LZ4LIB_API int LZ4_decompress_safe_usingDict (const char* src, char* dst, int sr
  * define LZ4_PUBLISH_STATIC_FUNCTIONS when building the LZ4 library.
  ******************************************************************************/
 
+#ifdef LZ4_STATIC_LINKING_ONLY
+
+#ifndef LZ4_STATIC_3504398509
+#define LZ4_STATIC_3504398509
+
 #ifdef LZ4_PUBLISH_STATIC_FUNCTIONS
 #define LZ4LIB_STATIC_API LZ4LIB_API
 #else
 #define LZ4LIB_STATIC_API
 #endif
 
-#ifdef LZ4_STATIC_LINKING_ONLY
-
 
 /*! LZ4_compress_fast_extState_fastReset() :
  *  A variant of LZ4_compress_fast_extState().
@@ -462,8 +476,75 @@ LZ4LIB_STATIC_API int LZ4_compress_fast_extState_fastReset (void* state, const c
  */
 LZ4LIB_STATIC_API void LZ4_attach_dictionary(LZ4_stream_t* workingStream, const LZ4_stream_t* dictionaryStream);
 
+
+/*! In-place compression and decompression
+ *
+ * It's possible to have input and output sharing the same buffer,
+ * for highly contrained memory environments.
+ * In both cases, it requires input to lay at the end of the buffer,
+ * and decompression to start at beginning of the buffer.
+ * Buffer size must feature some margin, hence be larger than final size.
+ *
+ * |<------------------------buffer--------------------------------->|
+ *                             |<-----------compressed data--------->|
+ * |<-----------decompressed size------------------>|
+ *                                                  |<----margin---->|
+ *
+ * This technique is more useful for decompression,
+ * since decompressed size is typically larger,
+ * and margin is short.
+ *
+ * In-place decompression will work inside any buffer
+ * which size is >= LZ4_DECOMPRESS_INPLACE_BUFFER_SIZE(decompressedSize).
+ * This presumes that decompressedSize > compressedSize.
+ * Otherwise, it means compression actually expanded data,
+ * and it would be more efficient to store such data with a flag indicating it's not compressed.
+ * This can happen when data is not compressible (already compressed, or encrypted).
+ *
+ * For in-place compression, margin is larger, as it must be able to cope with both
+ * history preservation, requiring input data to remain unmodified up to LZ4_DISTANCE_MAX,
+ * and data expansion, which can happen when input is not compressible.
+ * As a consequence, buffer size requirements are much higher,
+ * and memory savings offered by in-place compression are more limited.
+ *
+ * There are ways to limit this cost for compression :
+ * - Reduce history size, by modifying LZ4_DISTANCE_MAX.
+ *   Note that it is a compile-time constant, so all compressions will apply this limit.
+ *   Lower values will reduce compression ratio, except when input_size < LZ4_DISTANCE_MAX,
+ *   so it's a reasonable trick when inputs are known to be small.
+ * - Require the compressor to deliver a "maximum compressed size".
+ *   This is the `dstCapacity` parameter in `LZ4_compress*()`.
+ *   When this size is < LZ4_COMPRESSBOUND(inputSize), then compression can fail,
+ *   in which case, the return code will be 0 (zero).
+ *   The caller must be ready for these cases to happen,
+ *   and typically design a backup scheme to send data uncompressed.
+ * The combination of both techniques can significantly reduce
+ * the amount of margin required for in-place compression.
+ *
+ * In-place compression can work in any buffer
+ * which size is >= (maxCompressedSize)
+ * with maxCompressedSize == LZ4_COMPRESSBOUND(srcSize) for guaranteed compression success.
+ * LZ4_COMPRESS_INPLACE_BUFFER_SIZE() depends on both maxCompressedSize and LZ4_DISTANCE_MAX,
+ * so it's possible to reduce memory requirements by playing with them.
+ */
+
+#define LZ4_DECOMPRESS_INPLACE_MARGIN(compressedSize)          (((compressedSize) >> 8) + 32)
+#define LZ4_DECOMPRESS_INPLACE_BUFFER_SIZE(decompressedSize)   ((decompressedSize) + LZ4_DECOMPRESS_INPLACE_MARGIN(decompressedSize))  /**< note: presumes that compressedSize < decompressedSize. note2: margin is overestimated a bit, since it could use compressedSize instead */
+
+#ifndef LZ4_DISTANCE_MAX   /* history window size; can be user-defined at compile time */
+#  define LZ4_DISTANCE_MAX 65535   /* set to maximum value by default */
 #endif
 
+#define LZ4_COMPRESS_INPLACE_MARGIN                           (LZ4_DISTANCE_MAX + 32)   /* LZ4_DISTANCE_MAX can be safely replaced by srcSize when it's smaller */
+#define LZ4_COMPRESS_INPLACE_BUFFER_SIZE(maxCompressedSize)   ((maxCompressedSize) + LZ4_COMPRESS_INPLACE_MARGIN)  /**< maxCompressedSize is generally LZ4_COMPRESSBOUND(inputSize), but can be set to any lower value, with the risk that compression can fail (return code 0(zero)) */
+
+#endif   /* LZ4_STATIC_3504398509 */
+#endif   /* LZ4_STATIC_LINKING_ONLY */
+
+
+
+#ifndef LZ4_H_98237428734687
+#define LZ4_H_98237428734687
 
 /*-************************************************************
  *  PRIVATE DEFINITIONS
@@ -567,6 +648,7 @@ union LZ4_streamDecode_u {
 } ;   /* previously typedef'd to LZ4_streamDecode_t */
 
 
+
 /*-************************************
 *  Obsolete Functions
 **************************************/
@@ -601,8 +683,8 @@ union LZ4_streamDecode_u {
 #endif /* LZ4_DISABLE_DEPRECATE_WARNINGS */
 
 /* Obsolete compression functions */
-LZ4_DEPRECATED("use LZ4_compress_default() instead")       LZ4LIB_API int LZ4_compress               (const char* source, char* dest, int sourceSize);
-LZ4_DEPRECATED("use LZ4_compress_default() instead")       LZ4LIB_API int LZ4_compress_limitedOutput (const char* source, char* dest, int sourceSize, int maxOutputSize);
+LZ4_DEPRECATED("use LZ4_compress_default() instead")       LZ4LIB_API int LZ4_compress               (const char* src, char* dest, int srcSize);
+LZ4_DEPRECATED("use LZ4_compress_default() instead")       LZ4LIB_API int LZ4_compress_limitedOutput (const char* src, char* dest, int srcSize, int maxOutputSize);
 LZ4_DEPRECATED("use LZ4_compress_fast_extState() instead") LZ4LIB_API int LZ4_compress_withState               (void* state, const char* source, char* dest, int inputSize);
 LZ4_DEPRECATED("use LZ4_compress_fast_extState() instead") LZ4LIB_API int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize);
 LZ4_DEPRECATED("use LZ4_compress_fast_continue() instead") LZ4LIB_API int LZ4_compress_continue                (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize);
@@ -631,15 +713,18 @@ LZ4_DEPRECATED("use LZ4_decompress_safe_usingDict() instead") LZ4LIB_API int LZ4
 LZ4_DEPRECATED("use LZ4_decompress_fast_usingDict() instead") LZ4LIB_API int LZ4_decompress_fast_withPrefix64k (const char* src, char* dst, int originalSize);
 
 /*! LZ4_decompress_fast() : **unsafe!**
- *  These functions are generally slightly faster than LZ4_decompress_safe(),
- *  though the difference is small (generally ~5%).
- *  However, the real cost is a risk :  LZ4_decompress_safe() is protected vs malformed input, while `LZ4_decompress_fast()` is not, making it a security liability.
+ *  These functions used to be faster than LZ4_decompress_safe(),
+ *  but it has changed, and they are now slower than LZ4_decompress_safe().
+ *  This is because LZ4_decompress_fast() doesn't know the input size,
+ *  and therefore must progress more cautiously in the input buffer to not read beyond the end of block.
+ *  On top of that `LZ4_decompress_fast()` is not protected vs malformed or malicious inputs, making it a security liability.
  *  As a consequence, LZ4_decompress_fast() is strongly discouraged, and deprecated.
- *  These functions will generate a deprecation warning in the future.
  *
- *  Last LZ4_decompress_fast() specificity is that it can decompress a block without knowing its compressed size.
- *  Note that even that functionality could be achieved in a more secure manner if need be,
- *  though it would require new prototypes, and adaptation of the implementation to this new use case.
+ *  The last remaining LZ4_decompress_fast() specificity is that
+ *  it can decompress a block without knowing its compressed size.
+ *  Such functionality could be achieved in a more secure manner,
+ *  by also providing the maximum size of input buffer,
+ *  but it would require new prototypes, and adaptation of the implementation to this new use case.
  *
  *  Parameters:
  *  originalSize : is the uncompressed size to regenerate.
@@ -655,11 +740,11 @@ LZ4_DEPRECATED("use LZ4_decompress_fast_usingDict() instead") LZ4LIB_API int LZ4
  *         As a consequence, use these functions in trusted environments with trusted data **only**.
  */
 
-/* LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe() instead")  */
+LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe() instead")
 LZ4LIB_API int LZ4_decompress_fast (const char* src, char* dst, int originalSize);
-/* LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_continue() instead") */
+LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_continue() instead")
 LZ4LIB_API int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* src, char* dst, int originalSize);
-/* LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_usingDict() instead") */
+LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_usingDict() instead")
 LZ4LIB_API int LZ4_decompress_fast_usingDict (const char* src, char* dst, int originalSize, const char* dictStart, int dictSize);
 
 /*! LZ4_resetStream() :
@@ -671,7 +756,7 @@ LZ4LIB_API int LZ4_decompress_fast_usingDict (const char* src, char* dst, int or
 LZ4LIB_API void LZ4_resetStream (LZ4_stream_t* streamPtr);
 
 
-#endif /* LZ4_H_2983827168210 */
+#endif /* LZ4_H_98237428734687 */
 
 
 #if defined (__cplusplus)

data/contrib/lz4/lib/lz4frame.c

@@ -213,8 +213,8 @@ static void LZ4F_writeLE64 (void* dst, U64 value64)
 
 static const size_t minFHSize = LZ4F_HEADER_SIZE_MIN;   /*  7 */
 static const size_t maxFHSize = LZ4F_HEADER_SIZE_MAX;   /* 19 */
-static const size_t BHSize = 4;  /* block header : size, and compress flag */
-static const size_t BFSize = 4;  /* block footer : checksum (optional) */
+static const size_t BHSize = LZ4F_BLOCK_HEADER_SIZE;  /* block header : size, and compress flag */
+static const size_t BFSize = LZ4F_BLOCK_CHECKSUM_SIZE;  /* block footer : checksum (optional) */
 
 
 /*-************************************
@@ -325,9 +325,9 @@ static size_t LZ4F_compressBound_internal(size_t srcSize,
                                     const LZ4F_preferences_t* preferencesPtr,
                                           size_t alreadyBuffered)
 {
-    LZ4F_preferences_t prefsNull;
-    MEM_INIT(&prefsNull, 0, sizeof(prefsNull));
+    LZ4F_preferences_t prefsNull = LZ4F_INIT_PREFERENCES;
     prefsNull.frameInfo.contentChecksumFlag = LZ4F_contentChecksumEnabled;   /* worst case */
+    prefsNull.frameInfo.blockChecksumFlag = LZ4F_blockChecksumEnabled;   /* worst case */
     {   const LZ4F_preferences_t* const prefsPtr = (preferencesPtr==NULL) ? &prefsNull : preferencesPtr;
         U32 const flush = prefsPtr->autoFlush | (srcSize==0);
         LZ4F_blockSizeID_t const blockID = prefsPtr->frameInfo.blockSizeID;
@@ -1065,7 +1065,10 @@ struct LZ4F_dctx_s {
 LZ4F_errorCode_t LZ4F_createDecompressionContext(LZ4F_dctx** LZ4F_decompressionContextPtr, unsigned versionNumber)
 {
     LZ4F_dctx* const dctx = (LZ4F_dctx*)ALLOC_AND_ZERO(sizeof(LZ4F_dctx));
-    if (dctx==NULL) return err0r(LZ4F_ERROR_GENERIC);
+    if (dctx == NULL) {  /* failed allocation */
+        *LZ4F_decompressionContextPtr = NULL;
+        return err0r(LZ4F_ERROR_allocation_failed);
+    }
 
     dctx->version = versionNumber;
     *LZ4F_decompressionContextPtr = dctx;
@@ -1128,8 +1131,10 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
     }
 
     /* control magic number */
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
     if (LZ4F_readLE32(srcPtr) != LZ4F_MAGICNUMBER)
         return err0r(LZ4F_ERROR_frameType_unknown);
+#endif
     dctx->frameInfo.frameType = LZ4F_frame;
 
     /* Flags */
@@ -1168,10 +1173,12 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
 
     /* check header */
     assert(frameHeaderSize > 5);
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
     {   BYTE const HC = LZ4F_headerChecksum(srcPtr+4, frameHeaderSize-5);
         if (HC != srcPtr[frameHeaderSize-1])
             return err0r(LZ4F_ERROR_headerChecksum_invalid);
     }
+#endif
 
     /* save */
     dctx->frameInfo.blockMode = (LZ4F_blockMode_t)blockMode;
@@ -1208,8 +1215,10 @@ size_t LZ4F_headerSize(const void* src, size_t srcSize)
         return 8;
 
     /* control magic number */
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
     if (LZ4F_readLE32(src) != LZ4F_MAGICNUMBER)
         return err0r(LZ4F_ERROR_frameType_unknown);
+#endif
 
     /* Frame Header Size */
     {   BYTE const FLG = ((const BYTE*)src)[4];
@@ -1491,7 +1500,7 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
                 /* next block is a compressed block */
                 dctx->tmpInTarget = nextCBlockSize + crcSize;
                 dctx->dStage = dstage_getCBlock;
-                if (dstPtr==dstEnd) {
+                if (dstPtr==dstEnd || srcPtr==srcEnd) {
                     nextSrcSizeHint = BHSize + nextCBlockSize + crcSize;
                     doAnotherStage = 0;
                 }
@@ -1552,8 +1561,13 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
                 }
                 {   U32 const readCRC = LZ4F_readLE32(crcSrc);
                     U32 const calcCRC = XXH32_digest(&dctx->blockChecksum);
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
                     if (readCRC != calcCRC)
                         return err0r(LZ4F_ERROR_blockChecksum_invalid);
+#else
+                    (void)readCRC;
+                    (void)calcCRC;
+#endif
             }   }
             dctx->dStage = dstage_getBlockHeader;  /* new block */
             break;
@@ -1592,8 +1606,13 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
                 assert(selectedIn != NULL);  /* selectedIn is defined at this stage (either srcPtr, or dctx->tmpIn) */
                 {   U32 const readBlockCrc = LZ4F_readLE32(selectedIn + dctx->tmpInTarget);
                     U32 const calcBlockCrc = XXH32(selectedIn, dctx->tmpInTarget, 0);
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
                     if (readBlockCrc != calcBlockCrc)
                         return err0r(LZ4F_ERROR_blockChecksum_invalid);
+#else
+                    (void)readBlockCrc;
+                    (void)calcBlockCrc;
+#endif
             }   }
 
             if ((size_t)(dstEnd-dstPtr) >= dctx->maxBlockSize) {
@@ -1721,8 +1740,13 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
         /* case dstage_checkSuffix: */   /* no direct entry, avoid initialization risks */
             {   U32 const readCRC = LZ4F_readLE32(selectedIn);
                 U32 const resultCRC = XXH32_digest(&(dctx->xxh));
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
                 if (readCRC != resultCRC)
                     return err0r(LZ4F_ERROR_contentChecksum_invalid);
+#else
+                (void)readCRC;
+                (void)resultCRC;
+#endif
                 nextSrcSizeHint = 0;
                 LZ4F_resetDecompressionContext(dctx);
                 doAnotherStage = 0;

data/contrib/lz4/lib/lz4frame.h

@@ -176,7 +176,7 @@ typedef struct {
   LZ4F_blockChecksum_t   blockChecksumFlag;   /* 1: each block followed by a checksum of block's compressed data; 0: disabled (default) */
 } LZ4F_frameInfo_t;
 
-#define LZ4F_INIT_FRAMEINFO   { 0, 0, 0, 0, 0, 0, 0 }    /* v1.8.3+ */
+#define LZ4F_INIT_FRAMEINFO   { LZ4F_default, LZ4F_blockLinked, LZ4F_noContentChecksum, LZ4F_frame, 0ULL, 0U, LZ4F_noBlockChecksum }    /* v1.8.3+ */
 
 /*! LZ4F_preferences_t :
  *  makes it possible to supply advanced compression instructions to streaming interface.
@@ -191,7 +191,7 @@ typedef struct {
   unsigned reserved[3];         /* must be zero for forward compatibility */
 } LZ4F_preferences_t;
 
-#define LZ4F_INIT_PREFERENCES   { LZ4F_INIT_FRAMEINFO, 0, 0, 0, { 0, 0, 0 } }    /* v1.8.3+ */
+#define LZ4F_INIT_PREFERENCES   { LZ4F_INIT_FRAMEINFO, 0, 0u, 0u, { 0u, 0u, 0u } }    /* v1.8.3+ */
 
 
 /*-*********************************
@@ -253,6 +253,15 @@ LZ4FLIB_API LZ4F_errorCode_t LZ4F_freeCompressionContext(LZ4F_cctx* cctx);
 #define LZ4F_HEADER_SIZE_MIN  7   /* LZ4 Frame header size can vary, depending on selected paramaters */
 #define LZ4F_HEADER_SIZE_MAX 19
 
+/* Size in bytes of a block header in little-endian format. Highest bit indicates if block data is uncompressed */
+#define LZ4F_BLOCK_HEADER_SIZE 4
+
+/* Size in bytes of a block checksum footer in little-endian format. */
+#define LZ4F_BLOCK_CHECKSUM_SIZE 4
+
+/* Size in bytes of the content checksum. */
+#define LZ4F_CONTENT_CHECKSUM_SIZE 4
+
 /*! LZ4F_compressBegin() :
  *  will write the frame header into dstBuffer.
  *  dstCapacity must be >= LZ4F_HEADER_SIZE_MAX bytes.

data/contrib/lz4/lib/lz4hc.c

@@ -151,6 +151,21 @@ int LZ4HC_countBack(const BYTE* const ip, const BYTE* const match,
     return back;
 }
 
+#if defined(_MSC_VER)
+#  define LZ4HC_rotl32(x,r) _rotl(x,r)
+#else
+#  define LZ4HC_rotl32(x,r) ((x << r) | (x >> (32 - r)))
+#endif
+
+
+static U32 LZ4HC_rotatePattern(size_t const rotate, U32 const pattern)
+{
+    size_t const bitsToRotate = (rotate & (sizeof(pattern) - 1)) << 3;
+    if (bitsToRotate == 0)
+        return pattern;
+    return LZ4HC_rotl32(pattern, (int)bitsToRotate);
+}
+
 /* LZ4HC_countPattern() :
  * pattern32 must be a sample of repetitive pattern of length 1, 2 or 4 (but not 3!) */
 static unsigned
@@ -203,6 +218,16 @@ LZ4HC_reverseCountPattern(const BYTE* ip, const BYTE* const iLow, U32 pattern)
     return (unsigned)(iStart - ip);
 }
 
+/* LZ4HC_protectDictEnd() :
+ * Checks if the match is in the last 3 bytes of the dictionary, so reading the
+ * 4 byte MINMATCH would overflow.
+ * @returns true if the match index is okay.
+ */
+static int LZ4HC_protectDictEnd(U32 const dictLimit, U32 const matchIndex)
+{
+    return ((U32)((dictLimit - 1) - matchIndex) >= 3);
+}
+
 typedef enum { rep_untested, rep_not, rep_confirmed } repeat_state_e;
 typedef enum { favorCompressionRatio=0, favorDecompressionSpeed } HCfavor_e;
 
@@ -228,7 +253,7 @@ LZ4HC_InsertAndGetWiderMatch (
     const U32 dictLimit = hc4->dictLimit;
     const BYTE* const lowPrefixPtr = base + dictLimit;
     const U32 ipIndex = (U32)(ip - base);
-    const U32 lowestMatchIndex = (hc4->lowLimit + 64 KB > ipIndex) ? hc4->lowLimit : ipIndex - LZ4_DISTANCE_MAX;
+    const U32 lowestMatchIndex = (hc4->lowLimit + (LZ4_DISTANCE_MAX + 1) > ipIndex) ? hc4->lowLimit : ipIndex - LZ4_DISTANCE_MAX;
     const BYTE* const dictBase = hc4->dictBase;
     int const lookBackLength = (int)(ip-iLowLimit);
     int nbAttempts = maxNbAttempts;
@@ -287,14 +312,21 @@ LZ4HC_InsertAndGetWiderMatch (
         if (chainSwap && matchLength==longest) {    /* better match => select a better chain */
             assert(lookBackLength==0);   /* search forward only */
             if (matchIndex + (U32)longest <= ipIndex) {
+                int const kTrigger = 4;
                 U32 distanceToNextMatch = 1;
+                int const end = longest - MINMATCH + 1;
+                int step = 1;
+                int accel = 1 << kTrigger;
                 int pos;
-                for (pos = 0; pos <= longest - MINMATCH; pos++) {
+                for (pos = 0; pos < end; pos += step) {
                     U32 const candidateDist = DELTANEXTU16(chainTable, matchIndex + (U32)pos);
+                    step = (accel++ >> kTrigger);
                     if (candidateDist > distanceToNextMatch) {
                         distanceToNextMatch = candidateDist;
                         matchChainPos = (U32)pos;
-                }   }
+                        accel = 1 << kTrigger;
+                    }
+                }
                 if (distanceToNextMatch > 1) {
                     if (distanceToNextMatch > matchIndex) break;   /* avoid overflow */
                     matchIndex -= distanceToNextMatch;
@@ -313,34 +345,61 @@ LZ4HC_InsertAndGetWiderMatch (
                     } else {
                         repeat = rep_not;
                 }   }
-                if ( (repeat == rep_confirmed)
-                  && (matchCandidateIdx >= dictLimit) ) {   /* same segment only */
-                    const BYTE* const matchPtr = base + matchCandidateIdx;
+                if ( (repeat == rep_confirmed) && (matchCandidateIdx >= lowestMatchIndex)
+                  && LZ4HC_protectDictEnd(dictLimit, matchCandidateIdx) ) {
+                    const int extDict = matchCandidateIdx < dictLimit;
+                    const BYTE* const matchPtr = (extDict ? dictBase : base) + matchCandidateIdx;
                     if (LZ4_read32(matchPtr) == pattern) {  /* good candidate */
-                        size_t const forwardPatternLength = LZ4HC_countPattern(matchPtr+sizeof(pattern), iHighLimit, pattern) + sizeof(pattern);
-                        const BYTE* const lowestMatchPtr = (lowPrefixPtr + LZ4_DISTANCE_MAX >= ip) ? lowPrefixPtr : ip - LZ4_DISTANCE_MAX;
-                        size_t const backLength = LZ4HC_reverseCountPattern(matchPtr, lowestMatchPtr, pattern);
-                        size_t const currentSegmentLength = backLength + forwardPatternLength;
-
-                        if ( (currentSegmentLength >= srcPatternLength)   /* current pattern segment large enough to contain full srcPatternLength */
-                          && (forwardPatternLength <= srcPatternLength) ) { /* haven't reached this position yet */
-                            matchIndex = matchCandidateIdx + (U32)forwardPatternLength - (U32)srcPatternLength;  /* best position, full pattern, might be followed by more match */
-                        } else {
-                            matchIndex = matchCandidateIdx - (U32)backLength;   /* farthest position in current segment, will find a match of length currentSegmentLength + maybe some back */
-                            if (lookBackLength==0) {  /* no back possible */
-                                size_t const maxML = MIN(currentSegmentLength, srcPatternLength);
-                                if ((size_t)longest < maxML) {
-                                    assert(base + matchIndex < ip);
-                                    if (ip - (base+matchIndex) > LZ4_DISTANCE_MAX) break;
-                                    assert(maxML < 2 GB);
-                                    longest = (int)maxML;
-                                    *matchpos = base + matchIndex;   /* virtual pos, relative to ip, to retrieve offset */
-                                    *startpos = ip;
+                        const BYTE* const dictStart = dictBase + hc4->lowLimit;
+                        const BYTE* const iLimit = extDict ? dictBase + dictLimit : iHighLimit;
+                        size_t forwardPatternLength = LZ4HC_countPattern(matchPtr+sizeof(pattern), iLimit, pattern) + sizeof(pattern);
+                        if (extDict && matchPtr + forwardPatternLength == iLimit) {
+                            U32 const rotatedPattern = LZ4HC_rotatePattern(forwardPatternLength, pattern);
+                            forwardPatternLength += LZ4HC_countPattern(lowPrefixPtr, iHighLimit, rotatedPattern);
+                        }
+                        {   const BYTE* const lowestMatchPtr = extDict ? dictStart : lowPrefixPtr;
+                            size_t backLength = LZ4HC_reverseCountPattern(matchPtr, lowestMatchPtr, pattern);
+                            size_t currentSegmentLength;
+                            if (!extDict && matchPtr - backLength == lowPrefixPtr && hc4->lowLimit < dictLimit) {
+                                U32 const rotatedPattern = LZ4HC_rotatePattern((U32)(-(int)backLength), pattern);
+                                backLength += LZ4HC_reverseCountPattern(dictBase + dictLimit, dictStart, rotatedPattern);
+                            }
+                            /* Limit backLength not go further than lowestMatchIndex */
+                            backLength = matchCandidateIdx - MAX(matchCandidateIdx - (U32)backLength, lowestMatchIndex);
+                            assert(matchCandidateIdx - backLength >= lowestMatchIndex);
+                            currentSegmentLength = backLength + forwardPatternLength;
+                            /* Adjust to end of pattern if the source pattern fits, otherwise the beginning of the pattern */
+                            if ( (currentSegmentLength >= srcPatternLength)   /* current pattern segment large enough to contain full srcPatternLength */
+                              && (forwardPatternLength <= srcPatternLength) ) { /* haven't reached this position yet */
+                                U32 const newMatchIndex = matchCandidateIdx + (U32)forwardPatternLength - (U32)srcPatternLength;  /* best position, full pattern, might be followed by more match */
+                                if (LZ4HC_protectDictEnd(dictLimit, newMatchIndex))
+                                    matchIndex = newMatchIndex;
+                                else {
+                                    /* Can only happen if started in the prefix */
+                                    assert(newMatchIndex >= dictLimit - 3 && newMatchIndex < dictLimit && !extDict);
+                                    matchIndex = dictLimit;
                                 }
-                                {   U32 const distToNextPattern = DELTANEXTU16(chainTable, matchIndex);
-                                    if (distToNextPattern > matchIndex) break;  /* avoid overflow */
-                                    matchIndex -= distToNextPattern;
-                        }   }   }
+                            } else {
+                                U32 const newMatchIndex = matchCandidateIdx - (U32)backLength;   /* farthest position in current segment, will find a match of length currentSegmentLength + maybe some back */
+                                if (!LZ4HC_protectDictEnd(dictLimit, newMatchIndex)) {
+                                    assert(newMatchIndex >= dictLimit - 3 && newMatchIndex < dictLimit && !extDict);
+                                    matchIndex = dictLimit;
+                                } else {
+                                    matchIndex = newMatchIndex;
+                                    if (lookBackLength==0) {  /* no back possible */
+                                        size_t const maxML = MIN(currentSegmentLength, srcPatternLength);
+                                        if ((size_t)longest < maxML) {
+                                            assert(base + matchIndex < ip);
+                                            if (ip - (base+matchIndex) > LZ4_DISTANCE_MAX) break;
+                                            assert(maxML < 2 GB);
+                                            longest = (int)maxML;
+                                            *matchpos = base + matchIndex;   /* virtual pos, relative to ip, to retrieve offset */
+                                            *startpos = ip;
+                                        }
+                                        {   U32 const distToNextPattern = DELTANEXTU16(chainTable, matchIndex);
+                                            if (distToNextPattern > matchIndex) break;  /* avoid overflow */
+                                            matchIndex -= distToNextPattern;
+                        }   }   }   }   }
                         continue;
                 }   }
         }   }   /* PA optimization */
@@ -442,7 +501,7 @@ LZ4_FORCE_INLINE int LZ4HC_encodeSequence (
     }
 
     /* Copy Literals */
-    LZ4_wildCopy(*op, *anchor, (*op) + length);
+    LZ4_wildCopy8(*op, *anchor, (*op) + length);
     *op += length;
 
     /* Encode Offset */
@@ -1005,6 +1064,9 @@ static void LZ4HC_setExternalDict(LZ4HC_CCtx_internal* ctxPtr, const BYTE* newBl
     ctxPtr->base = newBlock - ctxPtr->dictLimit;
     ctxPtr->end  = newBlock;
     ctxPtr->nextToUpdate = ctxPtr->dictLimit;   /* match referencing will resume from there */
+
+    /* cannot reference an extDict and a dictCtx at the same time */
+    ctxPtr->dictCtx = NULL;
 }
 
 static int LZ4_compressHC_continue_generic (LZ4_streamHC_t* LZ4_streamHCPtr,
@@ -1396,6 +1458,7 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
              }   }
          }  /* for (cur = 1; cur <= last_match_pos; cur++) */
 
+         assert(last_match_pos < LZ4_OPT_NUM + TRAILING_LITERALS);
          best_mlen = opt[last_match_pos].mlen;
          best_off = opt[last_match_pos].off;
          cur = last_match_pos - best_mlen;