diff --git a/.gitignore b/.gitignore
index 01a4dd917..5f2ca97de 100644
--- a/.gitignore
+++ b/.gitignore
@@ -18,18 +18,15 @@ zstd
 *.out
 *.app
 
-# IDEA solution files
-*.idea
+# Test artefacts
+tmp*
+dictionary
 
 # Other files
 .directory
 _codelite/
 _zstdbench/
 .clang_complete
-
-# Test artefacts
-tmp*
-dictionary
-
-# tmp files
+*.idea
 *.swp
+.DS_Store
diff --git a/NEWS b/NEWS
index 36bd15242..70429644d 100644
--- a/NEWS
+++ b/NEWS
@@ -1,3 +1,7 @@
+v0.8.2
+API : ZDICT_getDictID()
+Small decompression speed improvement
+
 v0.8.1
 New streaming API
 Changed : --ultra now enables levels beyond 19
@@ -5,6 +9,7 @@ Changed : -i# now selects benchmark time in second
 Fixed : ZSTD_compress* can now compress > 4 GB in a single pass, reported by Nick Terrell
 Fixed : speed regression on specific patterns (#272)
 Fixed : support for Z_SYNC_FLUSH, by Dmitry Krot (#291)
+Fixed : ICC compilation, by Przemyslaw Skibinski
 
 v0.8.0
 Improved : better speed on clang and gcc -O2, thanks to Eric Biggers
diff --git a/lib/common/fse_decompress.c b/lib/common/fse_decompress.c
index 918de64c5..032e65771 100644
--- a/lib/common/fse_decompress.c
+++ b/lib/common/fse_decompress.c
@@ -68,6 +68,9 @@
 #define FSE_isError ERR_isError
 #define FSE_STATIC_ASSERT(c) { enum { FSE_static_assert = 1/(int)(!!(c)) }; }   /* use only *after* variable declarations */
 
+/* check and forward error code */
+#define CHECK_F(f) { size_t const e = f; if (FSE_isError(e)) return e; }
+
 
 /* **************************************************************
 *  Complex types
@@ -152,7 +155,6 @@ size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned
                 position = (position + step) & tableMask;
                 while (position > highThreshold) position = (position + step) & tableMask;   /* lowprob area */
         }   }
-
         if (position!=0) return ERROR(GENERIC);   /* position must reach all cells once, otherwise normalizedCounter is incorrect */
     }
 
@@ -169,7 +171,6 @@ size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned
 }
 
 
-
 #ifndef FSE_COMMONDEFS_ONLY
 
 /*-*******************************************************
@@ -234,8 +235,7 @@ FORCE_INLINE size_t FSE_decompress_usingDTable_generic(
     FSE_DState_t state2;
 
     /* Init */
-    { size_t const errorCode = BIT_initDStream(&bitD, cSrc, cSrcSize);   /* replaced last arg by maxCompressed Size */
-      if (FSE_isError(errorCode)) return errorCode; }
+    CHECK_F(BIT_initDStream(&bitD, cSrc, cSrcSize));
 
     FSE_initDState(&state1, &bitD, dt);
     FSE_initDState(&state2, &bitD, dt);
@@ -243,7 +243,7 @@ FORCE_INLINE size_t FSE_decompress_usingDTable_generic(
 #define FSE_GETSYMBOL(statePtr) fast ? FSE_decodeSymbolFast(statePtr, &bitD) : FSE_decodeSymbol(statePtr, &bitD)
 
     /* 4 symbols per loop */
-    for ( ; (BIT_reloadDStream(&bitD)==BIT_DStream_unfinished) && (op<olimit) ; op+=4) {
+    for ( ; (BIT_reloadDStream(&bitD)==BIT_DStream_unfinished) & (op<olimit) ; op+=4) {
         op[0] = FSE_GETSYMBOL(&state1);
 
         if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8)    /* This test must be static */
@@ -266,18 +266,14 @@ FORCE_INLINE size_t FSE_decompress_usingDTable_generic(
     /* note : BIT_reloadDStream(&bitD) >= FSE_DStream_partiallyFilled; Ends at exactly BIT_DStream_completed */
     while (1) {
         if (op>(omax-2)) return ERROR(dstSize_tooSmall);
-
         *op++ = FSE_GETSYMBOL(&state1);
-
         if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
             *op++ = FSE_GETSYMBOL(&state2);
             break;
         }
 
         if (op>(omax-2)) return ERROR(dstSize_tooSmall);
-
         *op++ = FSE_GETSYMBOL(&state2);
-
         if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
             *op++ = FSE_GETSYMBOL(&state1);
             break;
@@ -320,8 +316,7 @@ size_t FSE_decompress(void* dst, size_t maxDstSize, const void* cSrc, size_t cSr
         cSrcSize -= NCountLength;
     }
 
-    { size_t const errorCode = FSE_buildDTable (dt, counting, maxSymbolValue, tableLog);
-      if (FSE_isError(errorCode)) return errorCode; }
+    CHECK_F( FSE_buildDTable (dt, counting, maxSymbolValue, tableLog) );
 
     return FSE_decompress_usingDTable (dst, maxDstSize, ip, cSrcSize, dt);   /* always return, even if it is an error code */
 }
diff --git a/lib/compress/zstd_compress.c b/lib/compress/zstd_compress.c
index 00b1408ca..a696b2bc6 100644
--- a/lib/compress/zstd_compress.c
+++ b/lib/compress/zstd_compress.c
@@ -154,7 +154,7 @@ size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx)
     return 0;   /* reserved as a potential error code in the future */
 }
 
-size_t ZSTD_sizeofCCtx(const ZSTD_CCtx* cctx)
+size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx)
 {
     return sizeof(*cctx) + cctx->workSpaceSize;
 }
@@ -2867,6 +2867,10 @@ size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel)
     return ZSTD_initCStream_usingDict(zcs, NULL, 0, compressionLevel);
 }
 
+size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs)
+{
+    return sizeof(zcs) + ZSTD_sizeof_CCtx(zcs->zc) + zcs->outBuffSize + zcs->inBuffSize;
+}
 
 /*======   Compression   ======*/
 
diff --git a/lib/compress/zstd_opt.h b/lib/compress/zstd_opt.h
index f30cdbfda..612ac3b4d 100644
--- a/lib/compress/zstd_opt.h
+++ b/lib/compress/zstd_opt.h
@@ -529,7 +529,7 @@ void ZSTD_compressBlock_opt_generic(ZSTD_CCtx* ctx,
                continue;
 
            mlen = opt[cur].mlen;
-            if (opt[cur].off > ZSTD_REP_MOVE_OPT) {
+           if (opt[cur].off > ZSTD_REP_MOVE_OPT) {
                 opt[cur].rep[2] = opt[cur-mlen].rep[1];
                 opt[cur].rep[1] = opt[cur-mlen].rep[0];
                 opt[cur].rep[0] = opt[cur].off - ZSTD_REP_MOVE_OPT;
@@ -610,12 +610,12 @@ void ZSTD_compressBlock_opt_generic(ZSTD_CCtx* ctx,
                         price = opt[cur].price + ZSTD_getPrice(seqStorePtr, 0, NULL, matches[u].off-1, mlen - MINMATCH);
                     }
 
-                  //  ZSTD_LOG_PARSER("%d: Found2 mlen=%d best_mlen=%d off=%d price=%d litlen=%d\n", (int)(inr-base), mlen, best_mlen, matches[u].off, price, litlen);
+                    // ZSTD_LOG_PARSER("%d: Found2 mlen=%d best_mlen=%d off=%d price=%d litlen=%d\n", (int)(inr-base), mlen, best_mlen, matches[u].off, price, litlen);
                     if (cur + mlen > last_pos || (price < opt[cur + mlen].price))
                         SET_PRICE(cur + mlen, mlen, matches[u].off, litlen, price);
 
                     mlen++;
-        }   }   }   //  for (cur = 1; cur <= last_pos; cur++)
+        }   }   }   // for (cur = 1; cur <= last_pos; cur++)
 
         best_mlen = opt[last_pos].mlen;
         best_off = opt[last_pos].off;
diff --git a/lib/decompress/huf_decompress.c b/lib/decompress/huf_decompress.c
index 1580b3750..a5521bd36 100644
--- a/lib/decompress/huf_decompress.c
+++ b/lib/decompress/huf_decompress.c
@@ -37,7 +37,7 @@
 ****************************************************************/
 #if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
 /* inline is defined */
-#elif defined(_MSC_VER)
+#elif defined(_MSC_VER) || defined(__GNUC__)
 #  define inline __inline
 #else
 #  define inline /* disable inline */
@@ -57,10 +57,10 @@
 
 
 /* **************************************************************
-*  Includes
+*  Dependencies
 ****************************************************************/
 #include <string.h>     /* memcpy, memset */
-#include "bitstream.h"
+#include "bitstream.h"  /* BIT_* */
 #include "fse.h"        /* header compression */
 #define HUF_STATIC_LINKING_ONLY
 #include "huf.h"
@@ -103,7 +103,7 @@ size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize)
     HUF_DEltX2* const dt = (HUF_DEltX2*)dtPtr;
 
     HUF_STATIC_ASSERT(sizeof(DTableDesc) == sizeof(HUF_DTable));
-    //memset(huffWeight, 0, sizeof(huffWeight));   /* is not necessary, even though some analyzer complain ... */
+    /* memset(huffWeight, 0, sizeof(huffWeight)); */   /* is not necessary, even though some analyzer complain ... */
 
     iSize = HUF_readStats(huffWeight, HUF_SYMBOLVALUE_MAX + 1, rankVal, &nbSymbols, &tableLog, src, srcSize);
     if (HUF_isError(iSize)) return iSize;
@@ -388,22 +388,22 @@ static void HUF_fillDTableX4Level2(HUF_DEltX4* DTable, U32 sizeLog, const U32 co
     }
 
     /* fill DTable */
-    { U32 s; for (s=0; s<sortedListSize; s++) {   /* note : sortedSymbols already skipped */
-        const U32 symbol = sortedSymbols[s].symbol;
-        const U32 weight = sortedSymbols[s].weight;
-        const U32 nbBits = nbBitsBaseline - weight;
-        const U32 length = 1 << (sizeLog-nbBits);
-        const U32 start = rankVal[weight];
-        U32 i = start;
-        const U32 end = start + length;
+    {   U32 s; for (s=0; s<sortedListSize; s++) {   /* note : sortedSymbols already skipped */
+            const U32 symbol = sortedSymbols[s].symbol;
+            const U32 weight = sortedSymbols[s].weight;
+            const U32 nbBits = nbBitsBaseline - weight;
+            const U32 length = 1 << (sizeLog-nbBits);
+            const U32 start = rankVal[weight];
+            U32 i = start;
+            const U32 end = start + length;
 
-        MEM_writeLE16(&(DElt.sequence), (U16)(baseSeq + (symbol << 8)));
-        DElt.nbBits = (BYTE)(nbBits + consumed);
-        DElt.length = 2;
-        do { DTable[i++] = DElt; } while (i<end);   /* since length >= 1 */
+            MEM_writeLE16(&(DElt.sequence), (U16)(baseSeq + (symbol << 8)));
+            DElt.nbBits = (BYTE)(nbBits + consumed);
+            DElt.length = 2;
+            do { DTable[i++] = DElt; } while (i<end);   /* since length >= 1 */
 
-        rankVal[weight] += length;
-    }}
+            rankVal[weight] += length;
+    }   }
 }
 
 typedef U32 rankVal_t[HUF_TABLELOG_ABSOLUTEMAX][HUF_TABLELOG_ABSOLUTEMAX + 1];
@@ -442,8 +442,8 @@ static void HUF_fillDTableX4(HUF_DEltX4* DTable, const U32 targetLog,
             MEM_writeLE16(&(DElt.sequence), symbol);
             DElt.nbBits = (BYTE)(nbBits);
             DElt.length = 1;
-            {   U32 u;
-                const U32 end = start + length;
+            {   U32 const end = start + length;
+                U32 u;
                 for (u = start; u < end; u++) DTable[u] = DElt;
         }   }
         rankVal[weight] += length;
@@ -467,7 +467,7 @@ size_t HUF_readDTableX4 (HUF_DTable* DTable, const void* src, size_t srcSize)
 
     HUF_STATIC_ASSERT(sizeof(HUF_DEltX4) == sizeof(HUF_DTable));   /* if compilation fails here, assertion is false */
     if (maxTableLog > HUF_TABLELOG_ABSOLUTEMAX) return ERROR(tableLog_tooLarge);
-    //memset(weightList, 0, sizeof(weightList));   /* is not necessary, even though some analyzer complain ... */
+    /* memset(weightList, 0, sizeof(weightList)); */  /* is not necessary, even though some analyzer complain ... */
 
     iSize = HUF_readStats(weightList, HUF_SYMBOLVALUE_MAX + 1, rankStats, &nbSymbols, &tableLog, src, srcSize);
     if (HUF_isError(iSize)) return iSize;
@@ -533,7 +533,7 @@ size_t HUF_readDTableX4 (HUF_DTable* DTable, const void* src, size_t srcSize)
 
 static U32 HUF_decodeSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4* dt, const U32 dtLog)
 {
-    const size_t val = BIT_lookBitsFast(DStream, dtLog);   /* note : dtLog >= 1 */
+    size_t const val = BIT_lookBitsFast(DStream, dtLog);   /* note : dtLog >= 1 */
     memcpy(op, dt+val, 2);
     BIT_skipBits(DStream, dt[val].nbBits);
     return dt[val].length;
@@ -541,7 +541,7 @@ static U32 HUF_decodeSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4
 
 static U32 HUF_decodeLastSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4* dt, const U32 dtLog)
 {
-    const size_t val = BIT_lookBitsFast(DStream, dtLog);   /* note : dtLog >= 1 */
+    size_t const val = BIT_lookBitsFast(DStream, dtLog);   /* note : dtLog >= 1 */
     memcpy(op, dt+val, 1);
     if (dt[val].length==1) BIT_skipBits(DStream, dt[val].nbBits);
     else {
@@ -570,7 +570,7 @@ static inline size_t HUF_decodeStreamX4(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* c
     BYTE* const pStart = p;
 
     /* up to 8 symbols at a time */
-    while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p < pEnd-7)) {
+    while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-(sizeof(bitDPtr->bitContainer)-1))) {
         HUF_DECODE_SYMBOLX4_2(p, bitDPtr);
         HUF_DECODE_SYMBOLX4_1(p, bitDPtr);
         HUF_DECODE_SYMBOLX4_2(p, bitDPtr);
@@ -578,7 +578,7 @@ static inline size_t HUF_decodeStreamX4(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* c
     }
 
     /* closer to end : up to 2 symbols at a time */
-    while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p <= pEnd-2))
+    while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p <= pEnd-2))
         HUF_DECODE_SYMBOLX4_0(p, bitDPtr);
 
     while (p <= pEnd-2)
@@ -697,7 +697,7 @@ static size_t HUF_decompress4X4_usingDTable_internal(
 
         /* 16-32 symbols per loop (4-8 symbols per stream) */
         endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
-        for ( ; (endSignal==BIT_DStream_unfinished) && (op4<(oend-7)) ; ) {
+        for ( ; (endSignal==BIT_DStream_unfinished) & (op4<(oend-(sizeof(bitD4.bitContainer)-1))) ; ) {
             HUF_DECODE_SYMBOLX4_2(op1, &bitD1);
             HUF_DECODE_SYMBOLX4_2(op2, &bitD2);
             HUF_DECODE_SYMBOLX4_2(op3, &bitD3);
@@ -722,7 +722,7 @@ static size_t HUF_decompress4X4_usingDTable_internal(
         if (op1 > opStart2) return ERROR(corruption_detected);
         if (op2 > opStart3) return ERROR(corruption_detected);
         if (op3 > opStart4) return ERROR(corruption_detected);
-        /* note : op4 supposed already verified within main loop */
+        /* note : op4 already verified within main loop */
 
         /* finish bitStreams one by one */
         HUF_decodeStreamX4(op1, &bitD1, opStart2, dt, dtLog);
@@ -848,9 +848,6 @@ size_t HUF_decompress (void* dst, size_t dstSize, const void* cSrc, size_t cSrcS
     {   U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize);
         return decompress[algoNb](dst, dstSize, cSrc, cSrcSize);
     }
-
-    //return HUF_decompress4X2(dst, dstSize, cSrc, cSrcSize);   /* multi-streams single-symbol decoding */
-    //return HUF_decompress4X4(dst, dstSize, cSrc, cSrcSize);   /* multi-streams double-symbols decoding */
 }
 
 size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
diff --git a/lib/decompress/zstd_decompress.c b/lib/decompress/zstd_decompress.c
index 74e166add..524c2f56e 100644
--- a/lib/decompress/zstd_decompress.c
+++ b/lib/decompress/zstd_decompress.c
@@ -50,6 +50,16 @@
 #endif
 
 
+/*!
+*  STREAM_WINDOW_MAX :
+*  maximum window size accepted by DStream.
+*  frames requiring more memory will be rejected.
+*/
+#ifndef ZSTD_STREAM_WINDOW_MAX
+#  define ZSTD_STREAM_WINDOW_MAX (257 << 20)   /* 257 MB */
+#endif
+
+
 /*-*******************************************************
 *  Dependencies
 *********************************************************/
@@ -138,7 +148,7 @@ struct ZSTD_DCtx_s
     BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX];
 };  /* typedef'd to ZSTD_DCtx within "zstd_static.h" */
 
-size_t ZSTD_sizeofDCtx (const ZSTD_DCtx* dctx) { return sizeof(*dctx); }
+size_t ZSTD_sizeof_DCtx (const ZSTD_DCtx* dctx) { return sizeof(*dctx); }
 
 size_t ZSTD_estimateDCtxSize(void) { return sizeof(ZSTD_DCtx); }
 
@@ -153,7 +163,8 @@ size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx)
     dctx->hufTable[0] = (HUF_DTable)((HufLog)*0x1000001);
     dctx->litEntropy = dctx->fseEntropy = 0;
     dctx->dictID = 0;
-    { int i; for (i=0; i<ZSTD_REP_NUM; i++) dctx->rep[i] = repStartValue[i]; }
+    MEM_STATIC_ASSERT(sizeof(dctx->rep)==sizeof(repStartValue));
+    memcpy(dctx->rep, repStartValue, sizeof(repStartValue));
     return 0;
 }
 
@@ -161,15 +172,12 @@ ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem)
 {
     ZSTD_DCtx* dctx;
 
-    if (!customMem.customAlloc && !customMem.customFree)
-        customMem = defaultCustomMem;
-
-    if (!customMem.customAlloc || !customMem.customFree)
-        return NULL;
+    if (!customMem.customAlloc && !customMem.customFree) customMem = defaultCustomMem;
+    if (!customMem.customAlloc || !customMem.customFree) return NULL;
 
     dctx = (ZSTD_DCtx*) customMem.customAlloc(customMem.opaque, sizeof(ZSTD_DCtx));
     if (!dctx) return NULL;
-    memcpy(&dctx->customMem, &customMem, sizeof(ZSTD_customMem));
+    memcpy(&dctx->customMem, &customMem, sizeof(customMem));
     ZSTD_decompressBegin(dctx);
     return dctx;
 }
@@ -188,8 +196,8 @@ size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx)
 
 void ZSTD_copyDCtx(ZSTD_DCtx* dstDCtx, const ZSTD_DCtx* srcDCtx)
 {
-    memcpy(dstDCtx, srcDCtx,
-           sizeof(ZSTD_DCtx) - (ZSTD_BLOCKSIZE_ABSOLUTEMAX+WILDCOPY_OVERLENGTH + ZSTD_frameHeaderSize_max));  /* no need to copy workspace */
+    size_t const workSpaceSize = (ZSTD_BLOCKSIZE_ABSOLUTEMAX+WILDCOPY_OVERLENGTH) + ZSTD_frameHeaderSize_max;
+    memcpy(dstDCtx, srcDCtx, sizeof(ZSTD_DCtx) - workSpaceSize);  /* no need to copy workspace */
 }
 
 
@@ -210,7 +218,7 @@ static size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize)
         U32 const singleSegment = (fhd >> 5) & 1;
         U32 const fcsId = fhd >> 6;
         return ZSTD_frameHeaderSize_min + !singleSegment + ZSTD_did_fieldSize[dictID] + ZSTD_fcs_fieldSize[fcsId]
-                + (singleSegment && !ZSTD_fcs_fieldSize[fcsId]);
+                + (singleSegment && !fcsId);
     }
 }
 
@@ -592,9 +600,9 @@ static seq_t ZSTD_decodeSequence(seqState_t* seqState)
 {
     seq_t seq;
 
-    U32 const llCode = FSE_peekSymbol(&(seqState->stateLL));
-    U32 const mlCode = FSE_peekSymbol(&(seqState->stateML));
-    U32 const ofCode = FSE_peekSymbol(&(seqState->stateOffb));   /* <= maxOff, by table construction */
+    U32 const llCode = FSE_peekSymbol(&seqState->stateLL);
+    U32 const mlCode = FSE_peekSymbol(&seqState->stateML);
+    U32 const ofCode = FSE_peekSymbol(&seqState->stateOffb);   /* <= maxOff, by table construction */
 
     U32 const llBits = LL_bits[llCode];
     U32 const mlBits = ML_bits[mlCode];
@@ -623,8 +631,8 @@ static seq_t ZSTD_decodeSequence(seqState_t* seqState)
         if (!ofCode)
             offset = 0;
         else {
-            offset = OF_base[ofCode] + BIT_readBits(&(seqState->DStream), ofBits);   /* <=  (ZSTD_WINDOWLOG_MAX-1) bits */
-            if (MEM_32bits()) BIT_reloadDStream(&(seqState->DStream));
+            offset = OF_base[ofCode] + BIT_readBits(&seqState->DStream, ofBits);   /* <=  (ZSTD_WINDOWLOG_MAX-1) bits */
+            if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream);
         }
 
         if (ofCode <= 1) {
@@ -645,18 +653,18 @@ static seq_t ZSTD_decodeSequence(seqState_t* seqState)
         seq.offset = offset;
     }
 
-    seq.matchLength = ML_base[mlCode] + ((mlCode>31) ? BIT_readBits(&(seqState->DStream), mlBits) : 0);   /* <=  16 bits */
-    if (MEM_32bits() && (mlBits+llBits>24)) BIT_reloadDStream(&(seqState->DStream));
+    seq.matchLength = ML_base[mlCode] + ((mlCode>31) ? BIT_readBits(&seqState->DStream, mlBits) : 0);  /* <=  16 bits */
+    if (MEM_32bits() && (mlBits+llBits>24)) BIT_reloadDStream(&seqState->DStream);
 
-    seq.litLength = LL_base[llCode] + ((llCode>15) ? BIT_readBits(&(seqState->DStream), llBits) : 0);   /* <=  16 bits */
+    seq.litLength = LL_base[llCode] + ((llCode>15) ? BIT_readBits(&seqState->DStream, llBits) : 0);    /* <=  16 bits */
     if (MEM_32bits() ||
-       (totalBits > 64 - 7 - (LLFSELog+MLFSELog+OffFSELog)) ) BIT_reloadDStream(&(seqState->DStream));
+       (totalBits > 64 - 7 - (LLFSELog+MLFSELog+OffFSELog)) ) BIT_reloadDStream(&seqState->DStream);
 
     /* ANS state update */
-    FSE_updateState(&(seqState->stateLL), &(seqState->DStream));   /* <=  9 bits */
-    FSE_updateState(&(seqState->stateML), &(seqState->DStream));   /* <=  9 bits */
-    if (MEM_32bits()) BIT_reloadDStream(&(seqState->DStream));     /* <= 18 bits */
-    FSE_updateState(&(seqState->stateOffb), &(seqState->DStream)); /* <=  8 bits */
+    FSE_updateState(&seqState->stateLL, &seqState->DStream);    /* <=  9 bits */
+    FSE_updateState(&seqState->stateML, &seqState->DStream);    /* <=  9 bits */
+    if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream);    /* <= 18 bits */
+    FSE_updateState(&seqState->stateOffb, &seqState->DStream);  /* <=  8 bits */
 
     return seq;
 }
@@ -680,7 +688,9 @@ size_t ZSTD_execSequence(BYTE* op,
     if (iLitEnd > litLimit_w) return ERROR(corruption_detected);   /* over-read beyond lit buffer */
 
     /* copy Literals */
-    ZSTD_wildcopy(op, *litPtr, sequence.litLength);   /* note : since oLitEnd <= oend-WILDCOPY_OVERLENGTH, no risk of overwrite beyond oend */
+    ZSTD_copy8(op, *litPtr);
+    if (sequence.litLength > 8)
+        ZSTD_wildcopy(op+8, (*litPtr)+8, sequence.litLength - 8);   /* note : since oLitEnd <= oend-WILDCOPY_OVERLENGTH, no risk of overwrite beyond oend */
     op = oLitEnd;
     *litPtr = iLitEnd;   /* update for next sequence */
 
@@ -980,9 +990,10 @@ size_t ZSTD_decompress(void* dst, size_t dstCapacity, const void* src, size_t sr
 }
 
 
-/*-**********************************
-*   Streaming Decompression API
-************************************/
+/*-**************************************
+*   Advanced Streaming Decompression API
+*   Bufferless and synchronous
+****************************************/
 size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx) { return dctx->expected; }
 
 ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx) {
@@ -1382,6 +1393,11 @@ size_t ZSTD_initDStream(ZSTD_DStream* zds)
     return ZSTD_initDStream_usingDict(zds, NULL, 0);
 }
 
+size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds)
+{
+    return sizeof(*zds) + ZSTD_sizeof_DCtx(zds->zd) + zds->inBuffSize + zds->outBuffSize;
+}
+
 
 /* *** Decompression *** */
 
@@ -1439,6 +1455,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB
             /* Frame header instruct buffer sizes */
             {   size_t const blockSize = MIN(zds->fParams.windowSize, ZSTD_BLOCKSIZE_ABSOLUTEMAX);
                 size_t const neededOutSize = zds->fParams.windowSize + blockSize;
+                if (zds->fParams.windowSize > ZSTD_STREAM_WINDOW_MAX) return ERROR(frameParameter_unsupported);
                 zds->blockSize = blockSize;
                 if (zds->inBuffSize < blockSize) {
                     zds->customMem.customFree(zds->customMem.opaque, zds->inBuff);
@@ -1531,9 +1548,3 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB
         return nextSrcSizeHint;
     }
 }
-
-
-
-
-
-
diff --git a/lib/dictBuilder/zdict.c b/lib/dictBuilder/zdict.c
index 3416a979a..b2fb7c08f 100644
--- a/lib/dictBuilder/zdict.c
+++ b/lib/dictBuilder/zdict.c
@@ -126,6 +126,13 @@ unsigned ZDICT_isError(size_t errorCode) { return ERR_isError(errorCode); }
 
 const char* ZDICT_getErrorName(size_t errorCode) { return ERR_getErrorName(errorCode); }
 
+unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize)
+{
+    if (dictSize < 8) return 0;
+    if (MEM_readLE32(dictBuffer) != ZSTD_DICT_MAGIC) return 0;
+    return MEM_readLE32((const char*)dictBuffer + 4);
+}
+
 
 /*-********************************************************
 *  Dictionary training functions
@@ -692,7 +699,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     esr.workPlace = malloc(ZSTD_BLOCKSIZE_ABSOLUTEMAX);
     if (!esr.ref || !esr.zc || !esr.workPlace) {
         eSize = ERROR(memory_allocation);
-        DISPLAYLEVEL(1, "Not enough memory");
+        DISPLAYLEVEL(1, "Not enough memory \n");
         goto _cleanup;
     }
     if (offcodeMax>OFFCODE_MAX) { eSize = ERROR(dictionary_wrong); goto _cleanup; }   /* too large dictionary */
@@ -708,7 +715,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
 	{	size_t const beginResult = ZSTD_compressBegin_advanced(esr.ref, dictBuffer, dictBufferSize, params, 0);
 		if (ZSTD_isError(beginResult)) {
 			eSize = ERROR(GENERIC);
-			DISPLAYLEVEL(1, "error : ZSTD_compressBegin_advanced failed ");
+			DISPLAYLEVEL(1, "error : ZSTD_compressBegin_advanced failed \n");
 			goto _cleanup;
 	}	}
 
@@ -724,7 +731,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     errorCode = HUF_buildCTable (hufTable, countLit, 255, huffLog);
     if (HUF_isError(errorCode)) {
         eSize = ERROR(GENERIC);
-        DISPLAYLEVEL(1, "HUF_buildCTable error");
+        DISPLAYLEVEL(1, "HUF_buildCTable error \n");
         goto _cleanup;
     }
     huffLog = (U32)errorCode;
@@ -740,7 +747,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax);
     if (FSE_isError(errorCode)) {
         eSize = ERROR(GENERIC);
-        DISPLAYLEVEL(1, "FSE_normalizeCount error with offcodeCount");
+        DISPLAYLEVEL(1, "FSE_normalizeCount error with offcodeCount \n");
         goto _cleanup;
     }
     Offlog = (U32)errorCode;
@@ -749,7 +756,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML);
     if (FSE_isError(errorCode)) {
         eSize = ERROR(GENERIC);
-        DISPLAYLEVEL(1, "FSE_normalizeCount error with matchLengthCount");
+        DISPLAYLEVEL(1, "FSE_normalizeCount error with matchLengthCount \n");
         goto _cleanup;
     }
     mlLog = (U32)errorCode;
@@ -758,7 +765,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL);
     if (FSE_isError(errorCode)) {
         eSize = ERROR(GENERIC);
-        DISPLAYLEVEL(1, "FSE_normalizeCount error with litLengthCount");
+        DISPLAYLEVEL(1, "FSE_normalizeCount error with litLengthCount \n");
         goto _cleanup;
     }
     llLog = (U32)errorCode;
@@ -768,7 +775,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     {   size_t const hhSize = HUF_writeCTable(dstPtr, maxDstSize, hufTable, 255, huffLog);
         if (HUF_isError(hhSize)) {
             eSize = ERROR(GENERIC);
-            DISPLAYLEVEL(1, "HUF_writeCTable error");
+            DISPLAYLEVEL(1, "HUF_writeCTable error \n");
             goto _cleanup;
         }
         dstPtr += hhSize;
@@ -779,7 +786,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     {   size_t const ohSize = FSE_writeNCount(dstPtr, maxDstSize, offcodeNCount, OFFCODE_MAX, Offlog);
         if (FSE_isError(ohSize)) {
             eSize = ERROR(GENERIC);
-            DISPLAYLEVEL(1, "FSE_writeNCount error with offcodeNCount");
+            DISPLAYLEVEL(1, "FSE_writeNCount error with offcodeNCount \n");
             goto _cleanup;
         }
         dstPtr += ohSize;
@@ -790,7 +797,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     {   size_t const mhSize = FSE_writeNCount(dstPtr, maxDstSize, matchLengthNCount, MaxML, mlLog);
         if (FSE_isError(mhSize)) {
             eSize = ERROR(GENERIC);
-            DISPLAYLEVEL(1, "FSE_writeNCount error with matchLengthNCount");
+            DISPLAYLEVEL(1, "FSE_writeNCount error with matchLengthNCount \n");
             goto _cleanup;
         }
         dstPtr += mhSize;
@@ -801,7 +808,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     {   size_t const lhSize = FSE_writeNCount(dstPtr, maxDstSize, litLengthNCount, MaxLL, llLog);
         if (FSE_isError(lhSize)) {
             eSize = ERROR(GENERIC);
-            DISPLAYLEVEL(1, "FSE_writeNCount error with litlengthNCount");
+            DISPLAYLEVEL(1, "FSE_writeNCount error with litlengthNCount \n");
             goto _cleanup;
         }
         dstPtr += lhSize;
@@ -811,7 +818,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
 
     if (maxDstSize<12) {
         eSize = ERROR(GENERIC);
-        DISPLAYLEVEL(1, "not enough space to write RepOffsets");
+        DISPLAYLEVEL(1, "not enough space to write RepOffsets \n");
         goto _cleanup;
     }
 # if 0
@@ -856,10 +863,14 @@ size_t ZDICT_addEntropyTablesFromBuffer_advanced(void* dictBuffer, size_t dictCo
     /* entropy tables */
     DISPLAYLEVEL(2, "\r%70s\r", "");   /* clean display line */
     DISPLAYLEVEL(2, "statistics ... \n");
-    hSize += ZDICT_analyzeEntropy((char*)dictBuffer+hSize, dictBufferCapacity-hSize,
+    {   size_t const eSize = ZDICT_analyzeEntropy((char*)dictBuffer+hSize, dictBufferCapacity-hSize,
                                   compressionLevel,
                                   samplesBuffer, samplesSizes, nbSamples,
                                   (char*)dictBuffer + dictBufferCapacity - dictContentSize, dictContentSize);
+        if (ZDICT_isError(eSize)) return eSize;
+        hSize += eSize;
+    }
+
 
     if (hSize + dictContentSize < dictBufferCapacity)
         memmove((char*)dictBuffer + hSize, (char*)dictBuffer + dictBufferCapacity - dictContentSize, dictContentSize);
@@ -902,7 +913,7 @@ size_t ZDICT_trainFromBuffer_unsafe(
 
     /* display best matches */
     if (g_displayLevel>= 3) {
-        U32 const nb = 25;
+        U32 const nb = MIN(25, dictList[0].pos);
         U32 const dictContentSize = ZDICT_dictSize(dictList);
         U32 u;
         DISPLAYLEVEL(3, "\n %u segments found, of total size %u \n", dictList[0].pos, dictContentSize);
diff --git a/lib/dictBuilder/zdict.h b/lib/dictBuilder/zdict.h
index d61b5922c..68349da05 100644
--- a/lib/dictBuilder/zdict.h
+++ b/lib/dictBuilder/zdict.h
@@ -38,6 +38,19 @@
 extern "C" {
 #endif
 
+
+/*======  Export for Windows  ======*/
+/*!
+*  ZSTD_DLL_EXPORT :
+*  Enable exporting of functions when building a Windows DLL
+*/
+#if defined(_WIN32) && defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
+#  define ZDICTLIB_API __declspec(dllexport)
+#else
+#  define ZDICTLIB_API
+#endif
+
+
 /*! ZDICT_trainFromBuffer() :
     Train a dictionary from an array of samples.
     Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
@@ -50,13 +63,14 @@ extern "C" {
            In general, it's recommended to provide a few thousands samples, but this can vary a lot.
            It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
-size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
+ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
                        const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples);
 
 
 /*======   Helper functions   ======*/
-unsigned ZDICT_isError(size_t errorCode);
-const char* ZDICT_getErrorName(size_t errorCode);
+ZDICTLIB_API unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize);  /**< extracts dictID; @return zero if error (not a valid dictionary) */
+ZDICTLIB_API unsigned ZDICT_isError(size_t errorCode);
+ZDICTLIB_API const char* ZDICT_getErrorName(size_t errorCode);
 
 
 
diff --git a/lib/zstd.h b/lib/zstd.h
index 056940ff1..cb667484b 100644
--- a/lib/zstd.h
+++ b/lib/zstd.h
@@ -55,7 +55,7 @@ extern "C" {
 /*=======   Version   =======*/
 #define ZSTD_VERSION_MAJOR    0
 #define ZSTD_VERSION_MINOR    8
-#define ZSTD_VERSION_RELEASE  1
+#define ZSTD_VERSION_RELEASE  2
 
 #define ZSTD_LIB_VERSION ZSTD_VERSION_MAJOR.ZSTD_VERSION_MINOR.ZSTD_VERSION_RELEASE
 #define ZSTD_QUOTE(str) #str
@@ -92,7 +92,7 @@ ZSTDLIB_API size_t ZSTD_compress( void* dst, size_t dstCapacity,
 *             Always ensure result fits within application's authorized limits !
 *             Each application can set its own limits.
 *    note 4 : when `return==0`, if precise failure cause is needed, use ZSTD_getFrameParams() to know more. */
-unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize);
+ZSTDLIB_API unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize);
 
 /*! ZSTD_decompress() :
     `compressedSize` : must be the _exact_ size of compressed input, otherwise decompression will fail.
@@ -191,6 +191,107 @@ ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx,
                                         const ZSTD_DDict* ddict);
 
 
+/*-**************************
+*  Streaming
+****************************/
+
+typedef struct ZSTD_inBuffer_s {
+  const void* src;    /**< start of input buffer */
+  size_t size;        /**< size of input buffer */
+  size_t pos;         /**< position where reading stopped. Will be updated. Necessarily 0 <= pos <= size */
+} ZSTD_inBuffer;
+
+typedef struct ZSTD_outBuffer_s {
+  void*  dst;         /**< start of output buffer */
+  size_t size;        /**< size of output buffer */
+  size_t pos;         /**< position where writing stopped. Will be updated. Necessarily 0 <= pos <= size */
+} ZSTD_outBuffer;
+
+
+/*======   compression   ======*/
+
+/*-***********************************************************************
+*  Streaming compression - howto
+*
+*  A ZSTD_CStream object is required to track streaming operation.
+*  Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources.
+*  ZSTD_CStream objects can be reused multiple times on consecutive compression operations.
+*
+*  Start by initializing ZSTD_CStream.
+*  Use ZSTD_initCStream() to start a new compression operation.
+*  Use ZSTD_initCStream_usingDict() for a compression which requires a dictionary.
+*
+*  Use ZSTD_compressStream() repetitively to consume input stream.
+*  The function will automatically update both `pos`.
+*  Note that it may not consume the entire input, in which case `pos < size`,
+*  and it's up to the caller to present again remaining data.
+*  @return : a hint to preferred nb of bytes to use as input for next function call (it's just a hint, to improve latency)
+*            or an error code, which can be tested using ZSTD_isError().
+*
+*  At any moment, it's possible to flush whatever data remains within buffer, using ZSTD_flushStream().
+*  `output->pos` will be updated.
+*  Note some content might still be left within internal buffer if `output->size` is too small.
+*  @return : nb of bytes still present within internal buffer (0 if it's empty)
+*            or an error code, which can be tested using ZSTD_isError().
+*
+*  ZSTD_endStream() instructs to finish a frame.
+*  It will perform a flush and write frame epilogue.
+*  The epilogue is required for decoders to consider a frame completed.
+*  Similar to ZSTD_flushStream(), it may not be able to flush the full content if `output->size` is too small.
+*  In which case, call again ZSTD_endStream() to complete the flush.
+*  @return : nb of bytes still present within internal buffer (0 if it's empty)
+*            or an error code, which can be tested using ZSTD_isError().
+*
+* *******************************************************************/
+
+typedef struct ZSTD_CStream_s ZSTD_CStream;
+ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream(void);
+ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs);
+
+ZSTDLIB_API size_t ZSTD_CStreamInSize(void);    /**< recommended size for input buffer */
+ZSTDLIB_API size_t ZSTD_CStreamOutSize(void);   /**< recommended size for output buffer */
+
+ZSTDLIB_API size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel);
+ZSTDLIB_API size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
+ZSTDLIB_API size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
+ZSTDLIB_API size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
+
+
+/*======   decompression   ======*/
+
+/*-***************************************************************************
+*  Streaming decompression howto
+*
+*  A ZSTD_DStream object is required to track streaming operations.
+*  Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources.
+*  ZSTD_DStream objects can be re-init multiple times.
+*
+*  Use ZSTD_initDStream() to start a new decompression operation,
+*   or ZSTD_initDStream_usingDict() if decompression requires a dictionary.
+*
+*  Use ZSTD_decompressStream() repetitively to consume your input.
+*  The function will update both `pos`.
+*  Note that it may not consume the entire input (pos < size),
+*  in which case it's up to the caller to present remaining input again.
+*  @return : 0 when a frame is completely decoded and fully flushed,
+*            1 when there is still some data left within internal buffer to flush,
+*            >1 when more data is expected, with value being a suggested next input size (it's just a hint, which helps latency, any size is accepted),
+*            or an error code, which can be tested using ZSTD_isError().
+*
+* *******************************************************************************/
+
+typedef struct ZSTD_DStream_s ZSTD_DStream;
+ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream(void);
+ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds);
+
+ZSTDLIB_API size_t ZSTD_DStreamInSize(void);    /*!< recommended size for input buffer */
+ZSTDLIB_API size_t ZSTD_DStreamOutSize(void);   /*!< recommended size for output buffer */
+
+ZSTDLIB_API size_t ZSTD_initDStream(ZSTD_DStream* zds);
+ZSTDLIB_API size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
+
+
+
 #ifdef ZSTD_STATIC_LINKING_ONLY
 
 /* ====================================================================================
@@ -275,12 +376,12 @@ ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict_advanced(const void* dict, size_t dictS
 
 /*! ZSTD_sizeofCCtx() :
  *  Gives the amount of memory used by a given ZSTD_CCtx */
-ZSTDLIB_API size_t ZSTD_sizeofCCtx(const ZSTD_CCtx* cctx);
+ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx);
 
 /*! ZSTD_getParams() :
 *   same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of a `ZSTD_compressionParameters`.
 *   All fields of `ZSTD_frameParameters` are set to default (0) */
-ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long srcSize, size_t dictSize);
+ZSTDLIB_API ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long srcSize, size_t dictSize);
 
 /*! ZSTD_getCParams() :
 *   @return ZSTD_compressionParameters structure for a selected compression level and srcSize.
@@ -317,118 +418,28 @@ ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem);
 
 /*! ZSTD_sizeofDCtx() :
  *  Gives the amount of memory used by a given ZSTD_DCtx */
-ZSTDLIB_API size_t ZSTD_sizeofDCtx(const ZSTD_DCtx* dctx);
+ZSTDLIB_API size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx);
 
 
 /* ******************************************************************
-*  Streaming
+*  Advanced Streaming
 ********************************************************************/
 
-typedef struct ZSTD_inBuffer_s {
-  const void* src;    /**< start of input buffer */
-  size_t size;        /**< size of input buffer */
-  size_t pos;         /**< position where reading stopped. Will be updated. Necessarily 0 <= pos <= size */
-} ZSTD_inBuffer;
-
-typedef struct ZSTD_outBuffer_s {
-  void*  dst;         /**< start of output buffer */
-  size_t size;        /**< size of output buffer */
-  size_t pos;         /**< position where writing stopped. Will be updated. Necessarily 0 <= pos <= size */
-} ZSTD_outBuffer;
-
-
 /*======   compression   ======*/
 
-/*-***********************************************************************
-*  Streaming compression - howto
-*
-*  A ZSTD_CStream object is required to track streaming operation.
-*  Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources.
-*  ZSTD_CStream objects can be reused multiple times on consecutive compression operations.
-*
-*  Start by initializing ZSTD_CStream.
-*  Use ZSTD_initCStream() to start a new compression operation.
-*  Use ZSTD_initCStream_usingDict() for a compression which requires a dictionary.
-*
-*  Use ZSTD_compressStream() repetitively to consume input stream.
-*  The function will automatically update both `pos`.
-*  Note that it may not consume the entire input, in which case `pos < size`,
-*  and it's up to the caller to present again remaining data.
-*  @return : a hint to preferred nb of bytes to use as input for next function call (it's just a hint, to improve latency)
-*            or an error code, which can be tested using ZSTD_isError().
-*
-*  At any moment, it's possible to flush whatever data remains within buffer, using ZSTD_flushStream().
-*  `output->pos` will be updated.
-*  Note some content might still be left within internal buffer if `output->size` is too small.
-*  @return : nb of bytes still present within internal buffer (0 if it's empty)
-*            or an error code, which can be tested using ZSTD_isError().
-*
-*  ZSTD_endStream() instructs to finish a frame.
-*  It will perform a flush and write frame epilogue.
-*  The epilogue is required for decoders to consider a frame completed.
-*  Similar to ZSTD_flushStream(), it may not be able to flush the full content if `output->size` is too small.
-*  In which case, call again ZSTD_endStream() to complete the flush.
-*  @return : nb of bytes still present within internal buffer (0 if it's empty)
-*            or an error code, which can be tested using ZSTD_isError().
-*
-* *******************************************************************/
-
-typedef struct ZSTD_CStream_s ZSTD_CStream;
-ZSTD_CStream* ZSTD_createCStream(void);
-size_t ZSTD_freeCStream(ZSTD_CStream* zcs);
-
-size_t ZSTD_CStreamInSize(void);    /**< recommended size for input buffer */
-size_t ZSTD_CStreamOutSize(void);   /**< recommended size for output buffer */
-
-size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel);
-size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
-size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
-size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
-
-/* advanced */
-ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem);
-size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel);
-size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize,
+ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem);
+ZSTDLIB_API size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel);
+ZSTDLIB_API size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize,
                                  ZSTD_parameters params, unsigned long long pledgedSrcSize);
+ZSTDLIB_API size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs);
 
 
 /*======   decompression   ======*/
 
-/*-***************************************************************************
-*  Streaming decompression howto
-*
-*  A ZSTD_DStream object is required to track streaming operations.
-*  Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources.
-*  ZSTD_DStream objects can be re-init multiple times.
-*
-*  Use ZSTD_initDStream() to start a new decompression operation,
-*   or ZSTD_initDStream_usingDict() if decompression requires a dictionary.
-*
-*  Use ZSTD_decompressStream() repetitively to consume your input.
-*  The function will update both `pos`.
-*  Note that it may not consume the entire input (pos < size),
-*  in which case it's up to the caller to present remaining input again.
-*  @return : 0 when a frame is completely decoded and fully flushed,
-*            1 when there is still some data left within internal buffer to flush,
-*            >1 when more data is expected, with value being a suggested next input size (it's just a hint, which helps latency, any size is accepted),
-*            or an error code, which can be tested using ZSTD_isError().
-*
-* *******************************************************************************/
-
-typedef struct ZSTD_DStream_s ZSTD_DStream;
-ZSTD_DStream* ZSTD_createDStream(void);
-size_t ZSTD_freeDStream(ZSTD_DStream* zds);
-
-size_t ZSTD_DStreamInSize(void);    /*!< recommended size for input buffer */
-size_t ZSTD_DStreamOutSize(void);   /*!< recommended size for output buffer */
-
-size_t ZSTD_initDStream(ZSTD_DStream* zds);
-size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
-
 /* advanced */
-ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem);
-size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize);
-
+ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem);
+ZSTDLIB_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize);
+ZSTDLIB_API size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds);
 
 
 /* ******************************************************************
diff --git a/programs/bench.c b/programs/bench.c
index 12c59bff6..dc36d170f 100644
--- a/programs/bench.c
+++ b/programs/bench.c
@@ -257,7 +257,7 @@ static int BMK_benchMem(const void* srcBuffer, size_t srcSize,
             (void)fastestD; (void)crcOrig;   /*  unused when decompression disabled */
 #if 1
             /* Decompression */
-            memset(resultBuffer, 0xD6, srcSize);  /* warm result buffer */
+            if (!dCompleted) memset(resultBuffer, 0xD6, srcSize);  /* warm result buffer */
 
             UTIL_sleepMilli(1); /* give processor time to other processes */
             UTIL_waitForNextTick(ticksPerSecond);
diff --git a/programs/dibio.c b/programs/dibio.c
index cb864ec1d..19188d76f 100644
--- a/programs/dibio.c
+++ b/programs/dibio.c
@@ -204,11 +204,11 @@ int DiB_trainFromFiles(const char* dictFileName, unsigned maxDictSize,
     if ((!fileSizes) || (!srcBuffer) || (!dictBuffer)) EXM_THROW(12, "not enough memory for DiB_trainFiles");   /* should not happen */
     g_displayLevel = params.notificationLevel;
     if (nbFiles < 5) {
-        DISPLAYLEVEL(2, "!  Warning : nb of samples too low for proper processing \n");
-        DISPLAYLEVEL(2, "!  Please provide one file per sample \n");
-        DISPLAYLEVEL(2, "!  Avoid concatenating multiple samples into a single file \n");
-        DISPLAYLEVEL(2, "!  otherwise, dictBuilder will be unable to find the beginning of each sample \n");
-        DISPLAYLEVEL(2, "!  resulting in distorted statistics \n");
+        DISPLAYLEVEL(2, "!  Warning : nb of samples too low for proper processing ! \n");
+        DISPLAYLEVEL(2, "!  Please provide _one file per sample_. \n");
+        DISPLAYLEVEL(2, "!  Do not concatenate samples together into a single file, \n");
+        DISPLAYLEVEL(2, "!  as dictBuilder will be unable to find the beginning of each sample, \n");
+        DISPLAYLEVEL(2, "!  resulting in poor dictionary quality. \n");
     }
 
     /* init */
diff --git a/programs/fileio.c b/programs/fileio.c
index 397147739..c10c28968 100644
--- a/programs/fileio.c
+++ b/programs/fileio.c
@@ -364,9 +364,8 @@ static int FIO_compressFilename_internal(cRess_t ress,
     }
 
     /* Status */
-    if (strlen(srcFileName) > 20) srcFileName += strlen(srcFileName)-20; /* display last 20 characters */
     DISPLAYLEVEL(2, "\r%79s\r", "");
-    DISPLAYLEVEL(2,"%-20.20s :%6.2f%%   (%6llu => %6llu bytes, %s) \n", srcFileName,
+    DISPLAYLEVEL(2,"%-20s :%6.2f%%   (%6llu => %6llu bytes, %s) \n", srcFileName,
         (double)compressedfilesize/(readsize+(!readsize) /* avoid div by zero */ )*100,
         (unsigned long long)readsize, (unsigned long long) compressedfilesize,
          dstFileName);
diff --git a/tests/fuzzer.c b/tests/fuzzer.c
index be36d74bb..bed86f592 100644
--- a/tests/fuzzer.c
+++ b/tests/fuzzer.c
@@ -100,7 +100,6 @@ static unsigned FUZ_rand(unsigned* src)
     return rand32 >> 5;
 }
 
-
 static unsigned FUZ_highbit32(U32 v32)
 {
     unsigned nbBits = 0;
@@ -110,6 +109,10 @@ static unsigned FUZ_highbit32(U32 v32)
 }
 
 
+/*=============================================
+*   Basic Unit tests
+=============================================*/
+
 #define CHECK_V(var, fn)  size_t const var = fn; if (ZSTD_isError(var)) goto _output_error
 #define CHECK(fn)  { CHECK_V(err, fn); }
 #define CHECKPLUS(var, fn, more)  { CHECK_V(var, fn); more; }
@@ -272,6 +275,12 @@ static int basicUnitTests(U32 seed, double compressibility)
         if (ZDICT_isError(dictSize)) goto _output_error;
         DISPLAYLEVEL(4, "OK, created dictionary of size %u \n", (U32)dictSize);
 
+        DISPLAYLEVEL(4, "test%3i : check dictID : ", testNb++);
+        {   U32 const dictID = ZDICT_getDictID(dictBuffer, dictSize);
+            if (dictID==0) goto _output_error;
+            DISPLAYLEVEL(4, "OK : %u \n", dictID);
+        }
+
         DISPLAYLEVEL(4, "test%3i : compress with dictionary : ", testNb++);
         cSize = ZSTD_compress_usingDict(cctx, compressedBuffer, ZSTD_compressBound(CNBuffSize),
                                         CNBuffer, CNBuffSize,
diff --git a/tests/playTests.sh b/tests/playTests.sh
index 013ba6f64..75b334ff4 100755
--- a/tests/playTests.sh
+++ b/tests/playTests.sh
@@ -32,7 +32,7 @@ case "$OS" in
 esac
 
 MD5SUM="md5sum"
-if [ "$TRAVIS_OS_NAME" = "osx" ]; then
+if [[ "$OSTYPE" == "darwin"* ]]; then
     MD5SUM="md5 -r"
 fi
 
diff --git a/tests/zstreamtest.c b/tests/zstreamtest.c
index 7f136b1b5..4c81e8df6 100644
--- a/tests/zstreamtest.c
+++ b/tests/zstreamtest.c
@@ -61,7 +61,6 @@ static const U32 prime1 = 2654435761U;
 static const U32 prime2 = 2246822519U;
 
 
-
 /*-************************************
 *  Display Macros
 **************************************/
@@ -93,7 +92,6 @@ static U32 FUZ_GetMilliStart(void)
     return nCount;
 }
 
-
 static U32 FUZ_GetMilliSpan(U32 nTimeStart)
 {
     U32 const nCurrent = FUZ_GetMilliStart();
@@ -117,7 +115,6 @@ unsigned int FUZ_rand(unsigned int* seedPtr)
     return rand32 >> 5;
 }
 
-
 /*
 static unsigned FUZ_highbit32(U32 v32)
 {
@@ -141,6 +138,11 @@ static void freeFunction(void* opaque, void* address)
     free(address);
 }
 
+
+/*======================================================
+*   Basic Unit tests
+======================================================*/
+
 static int basicUnitTests(U32 seed, double compressibility, ZSTD_customMem customMem)
 {
     int testResult = 0;
@@ -187,6 +189,12 @@ static int basicUnitTests(U32 seed, double compressibility, ZSTD_customMem custo
     cSize += outBuff.pos;
     DISPLAYLEVEL(4, "OK (%u bytes : %.2f%%)\n", (U32)cSize, (double)cSize/COMPRESSIBLE_NOISE_LENGTH*100);
 
+    DISPLAYLEVEL(4, "test%3i : check CStream size : ", testNb++);
+    { size_t const s = ZSTD_sizeof_CStream(zc);
+      if (ZSTD_isError(s)) goto _output_error;
+      DISPLAYLEVEL(4, "OK (%u bytes) \n", (U32)s);
+    }
+
     /* skippable frame test */
     DISPLAYLEVEL(4, "test%3i : decompress skippable frame : ", testNb++);
     ZSTD_initDStream_usingDict(zd, CNBuffer, 128 KB);
@@ -218,6 +226,12 @@ static int basicUnitTests(U32 seed, double compressibility, ZSTD_customMem custo
     }   }
     DISPLAYLEVEL(4, "OK \n");
 
+    DISPLAYLEVEL(4, "test%3i : check DStream size : ", testNb++);
+    { size_t const s = ZSTD_sizeof_DStream(zd);
+      if (ZSTD_isError(s)) goto _output_error;
+      DISPLAYLEVEL(4, "OK (%u bytes) \n", (U32)s);
+    }
+
     /* Byte-by-byte decompression test */
     DISPLAYLEVEL(4, "test%3i : decompress byte-by-byte : ", testNb++);
     {   size_t r = 1;
diff --git a/zstd.rb b/zstd.rb
index 1817c2ebe..62b1e0b8f 100644
--- a/zstd.rb
+++ b/zstd.rb
@@ -1,8 +1,8 @@
 class Zstd < Formula
   desc "Zstandard - Fast real-time compression algorithm"
   homepage "http://www.zstd.net/"
-  url "https://github.com/Cyan4973/zstd/archive/v0.7.5.tar.gz"
-  sha256 "6800defac9a93ddb1d9673d62c78526800df71cf9f353456f8e488ba4de51061"
+  url "https://github.com/Cyan4973/zstd/archive/v0.8.1.tar.gz"
+  sha256 "4632bee45988dd0fe3edf1e67bdf0a833895cbb1a7d1eb23ef0b7d753f8bffdd"
 
   def install
     system "make", "install", "PREFIX=#{prefix}"
diff --git a/zstd_compression_format.md b/zstd_compression_format.md
index 0989912dd..61f6a57bf 100644
--- a/zstd_compression_format.md
+++ b/zstd_compression_format.md
@@ -443,7 +443,7 @@ using little-endian convention.
 
 In this representation, bits on the left are smallest bits.
 
-__`Literals_Block_Type`__ 
+__`Literals_Block_Type`__
 
 This field uses 2 lowest bits of first byte, describing 4 different block types :
 
@@ -460,7 +460,7 @@ This field uses 2 lowest bits of first byte, describing 4 different block types
         using Huffman tree _from previous Huffman-compressed literals block_.
         Huffman tree description will be skipped.
 
-__`Size_Format`__ 
+__`Size_Format`__
 
 `Size_Format` is divided into 2 families :
 
@@ -697,8 +697,13 @@ A match copy command specifies an offset and a length.
 The offset gives the position to copy from,
 which can be within a previous block.
 
-There are 3 symbol types, literals lengths, offsets and match lengths, 
-which are encoded together, interleaved in a single _bitstream_.
+When all _sequences_ are decoded,
+if there is any literal left in the _literal section_,
+these bytes are added at the end of the block.
+
+The _Sequences_Section_ regroup all symbols required to decode commands.
+There are 3 symbol types : literals lengths, offsets and match lengths.
+They are encoded together, interleaved, in a single _bitstream_.
 
 Each symbol is a _code_ in its own context,
 which specifies a baseline and a number of bits to add.
@@ -905,8 +910,8 @@ since it will be discovered and reported by the decoding process.
 
 The bitstream starts by reporting on which scale it operates.
 `Accuracy_Log = low4bits + 5`.
-Note that maximum `Accuracy_Log` for literal and match length is `9`,
-and for offsets it is `8`. Higher values are considered errors.
+Note that maximum `Accuracy_Log` for literal and match lengths is `9`,
+and for offsets is `8`. Higher values are considered errors.
 
 Then follow each symbol value, from `0` to last present one.
 The number of bits used by each field is variable.
@@ -1128,8 +1133,8 @@ _Reserved ranges :_
               However, for public distribution of compressed frames,
               the following ranges are reserved for future use and should not be used :
 
-              - low range : 1 - 32767 
-              - high range : >= (2^31) 
+              - low range : 1 - 32767
+              - high range : >= (2^31)
 
 __`Entropy_Tables`__ : following the same format as a [compressed blocks].
             They are stored in following order :