zstd-ruby 1.4.0.0 → 1.4.9.0

data/ext/zstdruby/libzstd/compress/zstd_double_fast.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * Copyright (c) 2016-2021, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -31,20 +31,19 @@ void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
      * is empty.
      */
     for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
-        U32 const current = (U32)(ip - base);
+        U32 const curr = (U32)(ip - base);
         U32 i;
         for (i = 0; i < fastHashFillStep; ++i) {
             size_t const smHash = ZSTD_hashPtr(ip + i, hBitsS, mls);
             size_t const lgHash = ZSTD_hashPtr(ip + i, hBitsL, 8);
             if (i == 0)
-                hashSmall[smHash] = current + i;
+                hashSmall[smHash] = curr + i;
             if (i == 0 || hashLarge[lgHash] == 0)
-                hashLarge[lgHash] = current + i;
+                hashLarge[lgHash] = curr + i;
             /* Only load extra positions for ZSTD_dtlm_full */
             if (dtlm == ZSTD_dtlm_fast)
                 break;
-        }
-    }
+    }   }
 }
 
 
@@ -63,7 +62,9 @@ size_t ZSTD_compressBlock_doubleFast_generic(
     const BYTE* const istart = (const BYTE*)src;
     const BYTE* ip = istart;
     const BYTE* anchor = istart;
-    const U32 prefixLowestIndex = ms->window.dictLimit;
+    const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
+    /* presumes that, if there is a dictionary, it must be using Attach mode */
+    const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
     const BYTE* const prefixLowest = base + prefixLowestIndex;
     const BYTE* const iend = istart + srcSize;
     const BYTE* const ilimit = iend - HASH_READ_SIZE;
@@ -93,14 +94,23 @@ size_t ZSTD_compressBlock_doubleFast_generic(
                                      dictCParams->hashLog : hBitsL;
     const U32 dictHBitsS           = dictMode == ZSTD_dictMatchState ?
                                      dictCParams->chainLog : hBitsS;
-    const U32 dictAndPrefixLength  = (U32)(ip - prefixLowest + dictEnd - dictStart);
+    const U32 dictAndPrefixLength  = (U32)((ip - prefixLowest) + (dictEnd - dictStart));
+
+    DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_generic");
 
     assert(dictMode == ZSTD_noDict || dictMode == ZSTD_dictMatchState);
 
+    /* if a dictionary is attached, it must be within window range */
+    if (dictMode == ZSTD_dictMatchState) {
+        assert(ms->window.dictLimit + (1U << cParams->windowLog) >= endIndex);
+    }
+
     /* init */
     ip += (dictAndPrefixLength == 0);
     if (dictMode == ZSTD_noDict) {
-        U32 const maxRep = (U32)(ip - prefixLowest);
+        U32 const curr = (U32)(ip - base);
+        U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog);
+        U32 const maxRep = curr - windowLow;
         if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
         if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
     }
@@ -119,17 +129,17 @@ size_t ZSTD_compressBlock_doubleFast_generic(
         size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
         size_t const dictHL = ZSTD_hashPtr(ip, dictHBitsL, 8);
         size_t const dictHS = ZSTD_hashPtr(ip, dictHBitsS, mls);
-        U32 const current = (U32)(ip-base);
+        U32 const curr = (U32)(ip-base);
         U32 const matchIndexL = hashLong[h2];
         U32 matchIndexS = hashSmall[h];
         const BYTE* matchLong = base + matchIndexL;
         const BYTE* match = base + matchIndexS;
-        const U32 repIndex = current + 1 - offset_1;
+        const U32 repIndex = curr + 1 - offset_1;
         const BYTE* repMatch = (dictMode == ZSTD_dictMatchState
                             && repIndex < prefixLowestIndex) ?
                                dictBase + (repIndex - dictIndexDelta) :
                                base + repIndex;
-        hashLong[h2] = hashSmall[h] = current;   /* update hash tables */
+        hashLong[h2] = hashSmall[h] = curr;   /* update hash tables */
 
         /* check dictMatchState repcode */
         if (dictMode == ZSTD_dictMatchState
@@ -138,7 +148,7 @@ size_t ZSTD_compressBlock_doubleFast_generic(
             const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
             mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
             goto _match_stored;
         }
 
@@ -147,7 +157,7 @@ size_t ZSTD_compressBlock_doubleFast_generic(
           && ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1)))) {
             mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
             goto _match_stored;
         }
 
@@ -167,11 +177,10 @@ size_t ZSTD_compressBlock_doubleFast_generic(
 
             if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) {
                 mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8;
-                offset = (U32)(current - dictMatchIndexL - dictIndexDelta);
+                offset = (U32)(curr - dictMatchIndexL - dictIndexDelta);
                 while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */
                 goto _match_found;
-            }
-        }
+        }   }
 
         if (matchIndexS > prefixLowestIndex) {
             /* check prefix short match */
@@ -186,20 +195,21 @@ size_t ZSTD_compressBlock_doubleFast_generic(
 
             if (match > dictStart && MEM_read32(match) == MEM_read32(ip)) {
                 goto _search_next_long;
-            }
-        }
+        }   }
 
         ip += ((ip-anchor) >> kSearchStrength) + 1;
+#if defined(__aarch64__)
+        PREFETCH_L1(ip+256);
+#endif
         continue;
 
 _search_next_long:
 
-        {
-            size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
+        {   size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
             size_t const dictHLNext = ZSTD_hashPtr(ip+1, dictHBitsL, 8);
             U32 const matchIndexL3 = hashLong[hl3];
             const BYTE* matchL3 = base + matchIndexL3;
-            hashLong[hl3] = current + 1;
+            hashLong[hl3] = curr + 1;
 
             /* check prefix long +1 match */
             if (matchIndexL3 > prefixLowestIndex) {
@@ -218,17 +228,15 @@ _search_next_long:
                 if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) {
                     mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8;
                     ip++;
-                    offset = (U32)(current + 1 - dictMatchIndexL3 - dictIndexDelta);
+                    offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta);
                     while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */
                     goto _match_found;
-                }
-            }
-        }
+        }   }   }
 
         /* if no long +1 match, explore the short match we found */
         if (dictMode == ZSTD_dictMatchState && matchIndexS < prefixLowestIndex) {
             mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4;
-            offset = (U32)(current - matchIndexS);
+            offset = (U32)(curr - matchIndexS);
             while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
         } else {
             mLength = ZSTD_count(ip+4, match+4, iend) + 4;
@@ -242,7 +250,7 @@ _match_found:
         offset_2 = offset_1;
         offset_1 = offset;
 
-        ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+        ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
 _match_stored:
         /* match found */
@@ -250,11 +258,14 @@ _match_stored:
         anchor = ip;
 
         if (ip <= ilimit) {
-            /* Fill Table */
-            hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] =
-                hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2;  /* here because current+2 could be > iend-8 */
-            hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] =
-                hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
+            /* Complementary insertion */
+            /* done after iLimit test, as candidates could be > iend-8 */
+            {   U32 const indexToInsert = curr+2;
+                hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
+                hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+                hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
+                hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
+            }
 
             /* check immediate repcode */
             if (dictMode == ZSTD_dictMatchState) {
@@ -263,14 +274,14 @@ _match_stored:
                     U32 const repIndex2 = current2 - offset_2;
                     const BYTE* repMatch2 = dictMode == ZSTD_dictMatchState
                         && repIndex2 < prefixLowestIndex ?
-                            dictBase - dictIndexDelta + repIndex2 :
+                            dictBase + repIndex2 - dictIndexDelta :
                             base + repIndex2;
                     if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
                        && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
                         const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
                         size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
                         U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
-                        ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
+                        ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
                         hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
                         hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
                         ip += repLength2;
@@ -278,8 +289,7 @@ _match_stored:
                         continue;
                     }
                     break;
-                }
-            }
+            }   }
 
             if (dictMode == ZSTD_noDict) {
                 while ( (ip <= ilimit)
@@ -290,18 +300,19 @@ _match_stored:
                     U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff;  /* swap offset_2 <=> offset_1 */
                     hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
                     hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
-                    ZSTD_storeSeq(seqStore, 0, anchor, 0, rLength-MINMATCH);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, rLength-MINMATCH);
                     ip += rLength;
                     anchor = ip;
                     continue;   /* faster when present ... (?) */
-    }   }   }   }
+        }   }   }
+    }   /* while (ip < ilimit) */
 
     /* save reps for next block */
     rep[0] = offset_1 ? offset_1 : offsetSaved;
     rep[1] = offset_2 ? offset_2 : offsetSaved;
 
     /* Return the last literals size */
-    return iend - anchor;
+    return (size_t)(iend - anchor);
 }
 
 
@@ -360,10 +371,13 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
     const BYTE* anchor = istart;
     const BYTE* const iend = istart + srcSize;
     const BYTE* const ilimit = iend - 8;
-    const U32   prefixStartIndex = ms->window.dictLimit;
     const BYTE* const base = ms->window.base;
+    const U32   endIndex = (U32)((size_t)(istart - base) + srcSize);
+    const U32   lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog);
+    const U32   dictStartIndex = lowLimit;
+    const U32   dictLimit = ms->window.dictLimit;
+    const U32   prefixStartIndex = (dictLimit > lowLimit) ? dictLimit : lowLimit;
     const BYTE* const prefixStart = base + prefixStartIndex;
-    const U32   dictStartIndex = ms->window.lowLimit;
     const BYTE* const dictBase = ms->window.dictBase;
     const BYTE* const dictStart = dictBase + dictStartIndex;
     const BYTE* const dictEnd = dictBase + prefixStartIndex;
@@ -371,6 +385,10 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
 
     DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_extDict_generic (srcSize=%zu)", srcSize);
 
+    /* if extDict is invalidated due to maxDistance, switch to "regular" variant */
+    if (prefixStartIndex == dictStartIndex)
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, mls, ZSTD_noDict);
+
     /* Search Loop */
     while (ip < ilimit) {  /* < instead of <=, because (ip+1) */
         const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls);
@@ -383,12 +401,12 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
         const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base;
         const BYTE* matchLong = matchLongBase + matchLongIndex;
 
-        const U32 current = (U32)(ip-base);
-        const U32 repIndex = current + 1 - offset_1;   /* offset_1 expected <= current +1 */
+        const U32 curr = (U32)(ip-base);
+        const U32 repIndex = curr + 1 - offset_1;   /* offset_1 expected <= curr +1 */
         const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
         const BYTE* const repMatch = repBase + repIndex;
         size_t mLength;
-        hashSmall[hSmall] = hashLong[hLong] = current;   /* update hash table */
+        hashSmall[hSmall] = hashLong[hLong] = curr;   /* update hash table */
 
         if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */
             & (repIndex > dictStartIndex))
@@ -396,18 +414,18 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
             const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
             mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
         } else {
             if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
                 const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
                 const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart;
                 U32 offset;
                 mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8;
-                offset = current - matchLongIndex;
+                offset = curr - matchLongIndex;
                 while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; }   /* catch up */
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
             } else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
                 size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
@@ -415,40 +433,44 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                 const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base;
                 const BYTE* match3 = match3Base + matchIndex3;
                 U32 offset;
-                hashLong[h3] = current + 1;
+                hashLong[h3] = curr + 1;
                 if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
                     const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend;
                     const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart;
                     mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8;
                     ip++;
-                    offset = current+1 - matchIndex3;
+                    offset = curr+1 - matchIndex3;
                     while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
                 } else {
                     const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
                     const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
                     mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
-                    offset = current - matchIndex;
+                    offset = curr - matchIndex;
                     while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; }   /* catch up */
                 }
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
             } else {
                 ip += ((ip-anchor) >> kSearchStrength) + 1;
                 continue;
         }   }
 
-        /* found a match : store it */
+        /* move to next sequence start */
         ip += mLength;
         anchor = ip;
 
         if (ip <= ilimit) {
-            /* Fill Table */
-            hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2;
-            hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] = current+2;
-            hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
-            hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+            /* Complementary insertion */
+            /* done after iLimit test, as candidates could be > iend-8 */
+            {   U32 const indexToInsert = curr+2;
+                hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
+                hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+                hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
+                hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
+            }
+
             /* check immediate repcode */
             while (ip <= ilimit) {
                 U32 const current2 = (U32)(ip-base);
@@ -460,7 +482,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                     const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
                     size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
                     U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
-                    ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
                     hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
                     hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
                     ip += repLength2;
@@ -475,7 +497,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
     rep[1] = offset_2;
 
     /* Return the last literals size */
-    return iend - anchor;
+    return (size_t)(iend - anchor);
 }
 
 

data/ext/zstdruby/libzstd/compress/zstd_double_fast.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * Copyright (c) 2016-2021, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -15,7 +15,7 @@
 extern "C" {
 #endif
 
-#include "mem.h"      /* U32 */
+#include "../common/mem.h"      /* U32 */
 #include "zstd_compress_internal.h"     /* ZSTD_CCtx, size_t */
 
 void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,

data/ext/zstdruby/libzstd/compress/zstd_fast.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * Copyright (c) 2016-2021, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -8,12 +8,13 @@
  * You may select, at your option, one of the above-listed licenses.
  */
 
-#include "zstd_compress_internal.h"
+#include "zstd_compress_internal.h"  /* ZSTD_hashPtr, ZSTD_count, ZSTD_storeSeq */
 #include "zstd_fast.h"
 
 
 void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
-                        void const* end, ZSTD_dictTableLoadMethod_e dtlm)
+                        const void* const end,
+                        ZSTD_dictTableLoadMethod_e dtlm)
 {
     const ZSTD_compressionParameters* const cParams = &ms->cParams;
     U32* const hashTable = ms->hashTable;
@@ -28,21 +29,22 @@ void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
      * Insert the other positions if their hash entry is empty.
      */
     for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) {
-        U32 const current = (U32)(ip - base);
+        U32 const curr = (U32)(ip - base);
         size_t const hash0 = ZSTD_hashPtr(ip, hBits, mls);
-        hashTable[hash0] = current;
+        hashTable[hash0] = curr;
         if (dtlm == ZSTD_dtlm_fast) continue;
         /* Only load extra positions for ZSTD_dtlm_full */
         {   U32 p;
             for (p = 1; p < fastHashFillStep; ++p) {
                 size_t const hash = ZSTD_hashPtr(ip + p, hBits, mls);
                 if (hashTable[hash] == 0) {  /* not yet filled */
-                    hashTable[hash] = current + p;
+                    hashTable[hash] = curr + p;
     }   }   }   }
 }
 
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_fast_generic(
+
+FORCE_INLINE_TEMPLATE size_t
+ZSTD_compressBlock_fast_generic(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize,
         U32 const mls)
@@ -58,7 +60,8 @@ size_t ZSTD_compressBlock_fast_generic(
     const BYTE* ip0 = istart;
     const BYTE* ip1;
     const BYTE* anchor = istart;
-    const U32   prefixStartIndex = ms->window.dictLimit;
+    const U32   endIndex = (U32)((size_t)(istart - base) + srcSize);
+    const U32   prefixStartIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
     const BYTE* const prefixStart = base + prefixStartIndex;
     const BYTE* const iend = istart + srcSize;
     const BYTE* const ilimit = iend - HASH_READ_SIZE;
@@ -66,15 +69,24 @@ size_t ZSTD_compressBlock_fast_generic(
     U32 offsetSaved = 0;
 
     /* init */
+    DEBUGLOG(5, "ZSTD_compressBlock_fast_generic");
     ip0 += (ip0 == prefixStart);
     ip1 = ip0 + 1;
-    {
-        U32 const maxRep = (U32)(ip0 - prefixStart);
+    {   U32 const curr = (U32)(ip0 - base);
+        U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog);
+        U32 const maxRep = curr - windowLow;
         if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
         if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
     }
 
     /* Main Search Loop */
+#ifdef __INTEL_COMPILER
+    /* From intel 'The vector pragma indicates that the loop should be
+     * vectorized if it is legal to do so'. Can be used together with
+     * #pragma ivdep (but have opted to exclude that because intel
+     * warns against using it).*/
+    #pragma vector always
+#endif
     while (ip1 < ilimit) {   /* < instead of <=, because check at ip0+2 */
         size_t mLength;
         BYTE const* ip2 = ip0 + 2;
@@ -86,19 +98,25 @@ size_t ZSTD_compressBlock_fast_generic(
         U32 const current1 = (U32)(ip1-base);
         U32 const matchIndex0 = hashTable[h0];
         U32 const matchIndex1 = hashTable[h1];
-        BYTE const* repMatch = ip2-offset_1;
+        BYTE const* repMatch = ip2 - offset_1;
         const BYTE* match0 = base + matchIndex0;
         const BYTE* match1 = base + matchIndex1;
         U32 offcode;
+
+#if defined(__aarch64__)
+        PREFETCH_L1(ip0+256);
+#endif
+
         hashTable[h0] = current0;   /* update hash table */
         hashTable[h1] = current1;   /* update hash table */
 
         assert(ip0 + 1 == ip1);
 
         if ((offset_1 > 0) & (MEM_read32(repMatch) == MEM_read32(ip2))) {
-            mLength = ip2[-1] == repMatch[-1] ? 1 : 0;
+            mLength = (ip2[-1] == repMatch[-1]) ? 1 : 0;
             ip0 = ip2 - mLength;
             match0 = repMatch - mLength;
+            mLength += 4;
             offcode = 0;
             goto _match;
         }
@@ -112,8 +130,7 @@ size_t ZSTD_compressBlock_fast_generic(
             match0 = match1;
             goto _offset;
         }
-        {
-            size_t const step = ((ip0-anchor) >> (kSearchStrength - 1)) + stepSize;
+        {   size_t const step = ((size_t)(ip0-anchor) >> (kSearchStrength - 1)) + stepSize;
             assert(step >= 2);
             ip0 += step;
             ip1 += step;
@@ -124,19 +141,18 @@ _offset: /* Requires: ip0, match0 */
         offset_2 = offset_1;
         offset_1 = (U32)(ip0-match0);
         offcode = offset_1 + ZSTD_REP_MOVE;
-        mLength = 0;
+        mLength = 4;
         /* Count the backwards match length */
         while (((ip0>anchor) & (match0>prefixStart))
              && (ip0[-1] == match0[-1])) { ip0--; match0--; mLength++; } /* catch up */
 
 _match: /* Requires: ip0, match0, offcode */
         /* Count the forward length */
-        mLength += ZSTD_count(ip0+mLength+4, match0+mLength+4, iend) + 4;
-        ZSTD_storeSeq(seqStore, ip0-anchor, anchor, offcode, mLength-MINMATCH);
+        mLength += ZSTD_count(ip0+mLength, match0+mLength, iend);
+        ZSTD_storeSeq(seqStore, (size_t)(ip0-anchor), anchor, iend, offcode, mLength-MINMATCH);
         /* match found */
         ip0 += mLength;
         anchor = ip0;
-        ip1 = ip0 + 1;
 
         if (ip0 <= ilimit) {
             /* Fill Table */
@@ -144,20 +160,18 @@ _match: /* Requires: ip0, match0, offcode */
             hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2;  /* here because current+2 could be > iend-8 */
             hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
 
-            while ( (ip0 <= ilimit)
-                 && ( (offset_2>0)
-                    & (MEM_read32(ip0) == MEM_read32(ip0 - offset_2)) )) {
-                /* store sequence */
-                size_t const rLength = ZSTD_count(ip0+4, ip0+4-offset_2, iend) + 4;
-                U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff;  /* swap offset_2 <=> offset_1 */
-                hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
-                ip0 += rLength;
-                ip1 = ip0 + 1;
-                ZSTD_storeSeq(seqStore, 0, anchor, 0, rLength-MINMATCH);
-                anchor = ip0;
-                continue;   /* faster when present (confirmed on gcc-8) ... (?) */
-            }
-        }
+            if (offset_2 > 0) { /* offset_2==0 means offset_2 is invalidated */
+                while ( (ip0 <= ilimit) && (MEM_read32(ip0) == MEM_read32(ip0 - offset_2)) ) {
+                    /* store sequence */
+                    size_t const rLength = ZSTD_count(ip0+4, ip0+4-offset_2, iend) + 4;
+                    { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
+                    hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
+                    ip0 += rLength;
+                    ZSTD_storeSeq(seqStore, 0 /*litLen*/, anchor, iend, 0 /*offCode*/, rLength-MINMATCH);
+                    anchor = ip0;
+                    continue;   /* faster when present (confirmed on gcc-8) ... (?) */
+        }   }   }
+        ip1 = ip0 + 1;
     }
 
     /* save reps for next block */
@@ -165,7 +179,7 @@ _match: /* Requires: ip0, match0, offcode */
     rep[1] = offset_2 ? offset_2 : offsetSaved;
 
     /* Return the last literals size */
-    return iend - anchor;
+    return (size_t)(iend - anchor);
 }
 
 
@@ -173,8 +187,7 @@ size_t ZSTD_compressBlock_fast(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize)
 {
-    ZSTD_compressionParameters const* cParams = &ms->cParams;
-    U32 const mls = cParams->minMatch;
+    U32 const mls = ms->cParams.minMatch;
     assert(ms->dictMatchState == NULL);
     switch(mls)
     {
@@ -222,11 +235,19 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
     const U32 dictAndPrefixLength  = (U32)(ip - prefixStart + dictEnd - dictStart);
     const U32 dictHLog             = dictCParams->hashLog;
 
-    /* otherwise, we would get index underflow when translating a dict index
-     * into a local index */
+    /* if a dictionary is still attached, it necessarily means that
+     * it is within window size. So we just check it. */
+    const U32 maxDistance = 1U << cParams->windowLog;
+    const U32 endIndex = (U32)((size_t)(ip - base) + srcSize);
+    assert(endIndex - prefixStartIndex <= maxDistance);
+    (void)maxDistance; (void)endIndex;   /* these variables are not used when assert() is disabled */
+
+    /* ensure there will be no underflow
+     * when translating a dict index into a local index */
     assert(prefixStartIndex >= (U32)(dictEnd - dictBase));
 
     /* init */
+    DEBUGLOG(5, "ZSTD_compressBlock_fast_dictMatchState_generic");
     ip += (dictAndPrefixLength == 0);
     /* dictMatchState repCode checks don't currently handle repCode == 0
      * disabling. */
@@ -237,21 +258,21 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
     while (ip < ilimit) {   /* < instead of <=, because repcode check at (ip+1) */
         size_t mLength;
         size_t const h = ZSTD_hashPtr(ip, hlog, mls);
-        U32 const current = (U32)(ip-base);
+        U32 const curr = (U32)(ip-base);
         U32 const matchIndex = hashTable[h];
         const BYTE* match = base + matchIndex;
-        const U32 repIndex = current + 1 - offset_1;
+        const U32 repIndex = curr + 1 - offset_1;
         const BYTE* repMatch = (repIndex < prefixStartIndex) ?
                                dictBase + (repIndex - dictIndexDelta) :
                                base + repIndex;
-        hashTable[h] = current;   /* update hash table */
+        hashTable[h] = curr;   /* update hash table */
 
         if ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex isn't overlapping dict + prefix */
           && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
             const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
             mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
         } else if ( (matchIndex <= prefixStartIndex) ) {
             size_t const dictHash = ZSTD_hashPtr(ip, dictHLog, mls);
             U32 const dictMatchIndex = dictHashTable[dictHash];
@@ -263,7 +284,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
                 continue;
             } else {
                 /* found a dict match */
-                U32 const offset = (U32)(current-dictMatchIndex-dictIndexDelta);
+                U32 const offset = (U32)(curr-dictMatchIndex-dictIndexDelta);
                 mLength = ZSTD_count_2segments(ip+4, dictMatch+4, iend, dictEnd, prefixStart) + 4;
                 while (((ip>anchor) & (dictMatch>dictStart))
                      && (ip[-1] == dictMatch[-1])) {
@@ -271,7 +292,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
                 } /* catch up */
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
             }
         } else if (MEM_read32(match) != MEM_read32(ip)) {
             /* it's not a match, and we're not going to check the dictionary */
@@ -286,7 +307,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
                  && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
             offset_2 = offset_1;
             offset_1 = offset;
-            ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
         }
 
         /* match found */
@@ -295,8 +316,8 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
 
         if (ip <= ilimit) {
             /* Fill Table */
-            assert(base+current+2 > istart);  /* check base overflow */
-            hashTable[ZSTD_hashPtr(base+current+2, hlog, mls)] = current+2;  /* here because current+2 could be > iend-8 */
+            assert(base+curr+2 > istart);  /* check base overflow */
+            hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2;  /* here because curr+2 could be > iend-8 */
             hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base);
 
             /* check immediate repcode */
@@ -311,7 +332,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
                     const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
                     size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
                     U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
-                    ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
                     hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2;
                     ip += repLength2;
                     anchor = ip;
@@ -327,15 +348,14 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
     rep[1] = offset_2 ? offset_2 : offsetSaved;
 
     /* Return the last literals size */
-    return iend - anchor;
+    return (size_t)(iend - anchor);
 }
 
 size_t ZSTD_compressBlock_fast_dictMatchState(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize)
 {
-    ZSTD_compressionParameters const* cParams = &ms->cParams;
-    U32 const mls = cParams->minMatch;
+    U32 const mls = ms->cParams.minMatch;
     assert(ms->dictMatchState != NULL);
     switch(mls)
     {
@@ -366,35 +386,46 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
     const BYTE* const istart = (const BYTE*)src;
     const BYTE* ip = istart;
     const BYTE* anchor = istart;
-    const U32   dictStartIndex = ms->window.lowLimit;
+    const U32   endIndex = (U32)((size_t)(istart - base) + srcSize);
+    const U32   lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog);
+    const U32   dictStartIndex = lowLimit;
     const BYTE* const dictStart = dictBase + dictStartIndex;
-    const U32   prefixStartIndex = ms->window.dictLimit;
+    const U32   dictLimit = ms->window.dictLimit;
+    const U32   prefixStartIndex = dictLimit < lowLimit ? lowLimit : dictLimit;
     const BYTE* const prefixStart = base + prefixStartIndex;
     const BYTE* const dictEnd = dictBase + prefixStartIndex;
     const BYTE* const iend = istart + srcSize;
     const BYTE* const ilimit = iend - 8;
     U32 offset_1=rep[0], offset_2=rep[1];
 
+    DEBUGLOG(5, "ZSTD_compressBlock_fast_extDict_generic (offset_1=%u)", offset_1);
+
+    /* switch to "regular" variant if extDict is invalidated due to maxDistance */
+    if (prefixStartIndex == dictStartIndex)
+        return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, mls);
+
     /* Search Loop */
     while (ip < ilimit) {  /* < instead of <=, because (ip+1) */
         const size_t h = ZSTD_hashPtr(ip, hlog, mls);
         const U32    matchIndex = hashTable[h];
         const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base;
         const BYTE*  match = matchBase + matchIndex;
-        const U32    current = (U32)(ip-base);
-        const U32    repIndex = current + 1 - offset_1;
+        const U32    curr = (U32)(ip-base);
+        const U32    repIndex = curr + 1 - offset_1;
         const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
         const BYTE* const repMatch = repBase + repIndex;
-        size_t mLength;
-        hashTable[h] = current;   /* update hash table */
-        assert(offset_1 <= current +1);   /* check repIndex */
+        hashTable[h] = curr;   /* update hash table */
+        DEBUGLOG(7, "offset_1 = %u , curr = %u", offset_1, curr);
+        assert(offset_1 <= curr +1);   /* check repIndex */
 
         if ( (((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > dictStartIndex))
            && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
-            const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
-            mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
+            const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
+            size_t const rLength = ZSTD_count_2segments(ip+1 +4, repMatch +4, iend, repMatchEnd, prefixStart) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, rLength-MINMATCH);
+            ip += rLength;
+            anchor = ip;
         } else {
             if ( (matchIndex < dictStartIndex) ||
                  (MEM_read32(match) != MEM_read32(ip)) ) {
@@ -402,36 +433,32 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
                 ip += ((ip-anchor) >> kSearchStrength) + stepSize;
                 continue;
             }
-            {   const BYTE* matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
-                const BYTE* lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
-                U32 offset;
-                mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
+            {   const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
+                const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
+                U32 const offset = curr - matchIndex;
+                size_t mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
                 while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; }   /* catch up */
-                offset = current - matchIndex;
-                offset_2 = offset_1;
-                offset_1 = offset;
-                ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                offset_2 = offset_1; offset_1 = offset;  /* update offset history */
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                ip += mLength;
+                anchor = ip;
         }   }
 
-        /* found a match : store it */
-        ip += mLength;
-        anchor = ip;
-
         if (ip <= ilimit) {
             /* Fill Table */
-            hashTable[ZSTD_hashPtr(base+current+2, hlog, mls)] = current+2;
+            hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2;
             hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base);
             /* check immediate repcode */
             while (ip <= ilimit) {
                 U32 const current2 = (U32)(ip-base);
                 U32 const repIndex2 = current2 - offset_2;
-                const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
+                const BYTE* const repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
                 if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (repIndex2 > dictStartIndex))  /* intentional overflow */
                    && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
                     const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
                     size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
-                    U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
-                    ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
+                    { U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; }  /* swap offset_2 <=> offset_1 */
+                    ZSTD_storeSeq(seqStore, 0 /*litlen*/, anchor, iend, 0 /*offcode*/, repLength2-MINMATCH);
                     hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2;
                     ip += repLength2;
                     anchor = ip;
@@ -445,7 +472,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
     rep[1] = offset_2;
 
     /* Return the last literals size */
-    return iend - anchor;
+    return (size_t)(iend - anchor);
 }
 
 
@@ -453,8 +480,7 @@ size_t ZSTD_compressBlock_fast_extDict(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize)
 {
-    ZSTD_compressionParameters const* cParams = &ms->cParams;
-    U32 const mls = cParams->minMatch;
+    U32 const mls = ms->cParams.minMatch;
     switch(mls)
     {
     default: /* includes case 3 */