From d605f482c7df515edf94a71ba762154c00735373 Mon Sep 17 00:00:00 2001 From: Yann Collet Date: Fri, 31 May 2019 16:50:04 -0700 Subject: [PATCH] make double_fast compatible with new index invalidation strategy --- lib/compress/zstd_double_fast.c | 41 ++++++++++++++++++++++----------- 1 file changed, 28 insertions(+), 13 deletions(-) diff --git a/lib/compress/zstd_double_fast.c b/lib/compress/zstd_double_fast.c index 47faf6d64..7bc2e3fa0 100644 --- a/lib/compress/zstd_double_fast.c +++ b/lib/compress/zstd_double_fast.c @@ -43,8 +43,7 @@ void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms, /* Only load extra positions for ZSTD_dtlm_full */ if (dtlm == ZSTD_dtlm_fast) break; - } - } + } } } @@ -63,7 +62,10 @@ 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); + const U32 lowestValid = ms->window.dictLimit; + const U32 maxDistance = 1U << cParams->windowLog; + const U32 prefixLowestIndex = (endIndex - lowestValid > maxDistance) ? endIndex - maxDistance : lowestValid; const BYTE* const prefixLowest = base + prefixLowestIndex; const BYTE* const iend = istart + srcSize; const BYTE* const ilimit = iend - HASH_READ_SIZE; @@ -97,6 +99,10 @@ size_t 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(lowestValid + maxDistance <= endIndex); + /* init */ ip += (dictAndPrefixLength == 0); if (dictMode == ZSTD_noDict) { @@ -138,7 +144,7 @@ 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, 0, mLength-MINMATCH); goto _match_stored; } @@ -147,7 +153,7 @@ 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, 0, mLength-MINMATCH); goto _match_stored; } @@ -242,7 +248,7 @@ _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, offset + ZSTD_REP_MOVE, mLength-MINMATCH); _match_stored: /* match found */ @@ -301,7 +307,7 @@ _match_stored: rep[1] = offset_2 ? offset_2 : offsetSaved; /* Return the last literals size */ - return iend - anchor; + return (size_t)(iend - anchor); } @@ -360,10 +366,15 @@ 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 maxDistance = 1U << cParams->windowLog; + const U32 lowestValid = ms->window.lowLimit; + const U32 lowLimit = (endIndex - lowestValid > maxDistance) ? endIndex - maxDistance : lowestValid; + 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 +382,10 @@ 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); @@ -396,7 +411,7 @@ 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, 0, mLength-MINMATCH); } else { if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) { const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend; @@ -407,7 +422,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic( 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, 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); @@ -432,7 +447,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic( } 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, offset + ZSTD_REP_MOVE, mLength-MINMATCH); } else { ip += ((ip-anchor) >> kSearchStrength) + 1; @@ -475,7 +490,7 @@ 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); } -- 2.47.2