summaryrefslogtreecommitdiffstats
path: root/lib/lz4hc.c
diff options
context:
space:
mode:
authorYann Collet <cyan@fb.com>2017-10-09 06:55:42 (GMT)
committerYann Collet <cyan@fb.com>2017-10-09 06:55:42 (GMT)
commitb9459faeb24afe8669640aef8861ce32f73d5aae (patch)
tree161f29cc2e415c7da1b67ab9ed81ef193218a234 /lib/lz4hc.c
parentf1fa91d6fc74576dcde98b6514cfc8cc57f6f50f (diff)
downloadlz4-b9459faeb24afe8669640aef8861ce32f73d5aae.zip
lz4-b9459faeb24afe8669640aef8861ce32f73d5aae.tar.gz
lz4-b9459faeb24afe8669640aef8861ce32f73d5aae.tar.bz2
improved search of rep-1 patterns
Diffstat (limited to 'lib/lz4hc.c')
-rw-r--r--lib/lz4hc.c158
1 files changed, 131 insertions, 27 deletions
diff --git a/lib/lz4hc.c b/lib/lz4hc.c
index f4a0981..e28d682 100644
--- a/lib/lz4hc.c
+++ b/lib/lz4hc.c
@@ -165,6 +165,54 @@ LZ4_FORCE_INLINE int LZ4HC_InsertAndFindBestMatch (LZ4HC_CCtx_internal* const hc
}
#endif
+/** LZ4HC_countBack() :
+ * @return : negative value, nb of common bytes before ip/match */
+static int LZ4HC_countBack(const BYTE* const ip, const BYTE* const match,
+ const BYTE* const iMin, const BYTE* const mMin)
+{
+ int back=0;
+ while ( (ip+back > iMin)
+ && (match+back > mMin)
+ && (ip[back-1] == match[back-1]))
+ back--;
+ return back;
+}
+
+static unsigned LZ4HC_countPattern(const BYTE* ip, const BYTE* const iEnd, reg_t pattern)
+{
+ const BYTE* const iStart = ip;
+
+ while (likely(ip<iEnd-(sizeof(pattern)-1))) {
+ reg_t const diff = LZ4_read_ARCH(ip) ^ pattern;
+ if (!diff) { ip+=sizeof(pattern); continue; }
+ ip += LZ4_NbCommonBytes(diff);
+ return (unsigned)(ip - iStart);
+ }
+
+ if ((sizeof(pattern)==8) && (ip<(iEnd-3)) && (LZ4_read32(ip)==(U32)pattern)) { ip+=4; }
+ if ((ip<(iEnd-1)) && (LZ4_read16(ip)==(U16)pattern)) { ip+=2; }
+ if ((ip<iEnd) && (*ip == (BYTE)pattern)) ip++;
+ return (unsigned)(ip - iStart);
+}
+
+static unsigned LZ4HC_reverseCountPattern(const BYTE* ip, const BYTE* const iLow, U32 pattern)
+{
+ const BYTE* const iStart = ip;
+
+ while (likely(ip>=iLow+4)) {
+ if (LZ4_read32(ip-4) != pattern) break;
+ ip -= 4;
+ }
+ while (likely(ip>iLow)) {
+ if (ip[-1] != (BYTE)pattern) break;
+ ip--;
+ }
+
+ return (unsigned)(iStart - ip);
+}
+
+typedef enum { rep_untested, rep_not, rep_confirmed } repeat_state_e;
+
LZ4_FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (
LZ4HC_CCtx_internal* hc4,
const BYTE* const ip,
@@ -180,11 +228,13 @@ LZ4_FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (
const BYTE* const base = hc4->base;
const U32 dictLimit = hc4->dictLimit;
const BYTE* const lowPrefixPtr = base + dictLimit;
- const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1);
+ const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - MAX_DISTANCE;
const BYTE* const dictBase = hc4->dictBase;
- int const delta = (int)(ip-iLowLimit);
int nbAttempts = maxNbAttempts;
+ reg_t const pattern = LZ4_read_ARCH(ip);
U32 matchIndex;
+ repeat_state_e repeat = rep_untested;
+ size_t srcPatternLength = 0;
/* First Match */
@@ -195,27 +245,29 @@ LZ4_FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (
nbAttempts--;
if (matchIndex >= dictLimit) {
const BYTE* const matchPtr = base + matchIndex;
- if (*(iLowLimit + longest) == *(matchPtr - delta + longest)) {
- if (LZ4_read32(matchPtr) == LZ4_read32(ip)) {
- int mlt = MINMATCH + LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, iHighLimit);
- int back = 0;
-
- while ( (ip+back > iLowLimit)
- && (matchPtr+back > lowPrefixPtr)
- && (ip[back-1] == matchPtr[back-1])) {
- back--;
- }
-
- mlt -= back;
+ if (LZ4_read32(matchPtr) == (U32)pattern) {
+ int mlt = MINMATCH + LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, iHighLimit);
+#if 0
+ /* more generic but unfortunately slower ... */
+ int const back = LZ4HC_countBack(ip, matchPtr, iLowLimit, lowPrefixPtr);
+#else
+ int back = 0;
+ while ( (ip+back > iLowLimit)
+ && (matchPtr+back > lowPrefixPtr)
+ && (ip[back-1] == matchPtr[back-1])) {
+ back--;
+ }
+#endif
+ mlt -= back;
- if (mlt > longest) {
- longest = mlt;
- *matchpos = matchPtr+back;
- *startpos = ip+back;
- } } }
- } else {
+ if (mlt > longest) {
+ longest = mlt;
+ *matchpos = matchPtr+back;
+ *startpos = ip+back;
+ } }
+ } else { /* matchIndex < dictLimit */
const BYTE* const matchPtr = dictBase + matchIndex;
- if (LZ4_read32(matchPtr) == LZ4_read32(ip)) {
+ if (LZ4_read32(matchPtr) == (U32)pattern) {
int mlt;
int back=0;
const BYTE* vLimit = ip + (dictLimit - matchIndex);
@@ -223,13 +275,65 @@ LZ4_FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (
mlt = LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH;
if ((ip+mlt == vLimit) && (vLimit < iHighLimit))
mlt += LZ4_count(ip+mlt, base+dictLimit, iHighLimit);
- while ((ip+back > iLowLimit) && (matchIndex+back > lowLimit) && (ip[back-1] == matchPtr[back-1])) back--;
+ while ( (ip+back > iLowLimit)
+ && (matchIndex+back > lowLimit)
+ && (ip[back-1] == matchPtr[back-1]))
+ back--;
mlt -= back;
- if (mlt > longest) { longest = mlt; *matchpos = base + matchIndex + back; *startpos = ip+back; }
- }
- }
- matchIndex -= DELTANEXTU16(chainTable, matchIndex);
- }
+ if (mlt > longest) {
+ longest = mlt;
+ *matchpos = base + matchIndex + back;
+ *startpos = ip + back;
+ } } }
+
+ { U32 const nextOffset = DELTANEXTU16(chainTable, matchIndex);
+ matchIndex -= nextOffset;
+ if (1 && (nextOffset==1)) {
+ /* may be a repeated pattern */
+ if (repeat == rep_untested) {
+ if (LZ4_read32(ip+4) == (U32)pattern) { /* should check ip limit */
+ repeat = rep_confirmed;
+ srcPatternLength = LZ4HC_countPattern(ip+8, iHighLimit, pattern) + 8;
+ } else {
+ repeat = rep_not;
+ } }
+ if ( (repeat == rep_confirmed) /* proven repeated pattern (1-2-4) */
+ && (matchIndex >= dictLimit) ) { /* same segment only */
+ const BYTE* const matchPtr = base + matchIndex;
+ if (LZ4_read_ARCH(matchPtr) == pattern) { /* good candidate */
+ size_t const forwardPatternLength = LZ4HC_countPattern(matchPtr+sizeof(pattern), iHighLimit, pattern) + sizeof(pattern);
+ const BYTE* const maxLowPtr = (lowPrefixPtr + MAX_DISTANCE >= ip) ? lowPrefixPtr : ip - MAX_DISTANCE;
+ size_t const backLength = LZ4HC_reverseCountPattern(matchPtr, maxLowPtr, (U32)pattern);
+ size_t const currentSegmentLength = backLength + forwardPatternLength;
+
+ if ( (currentSegmentLength >= srcPatternLength) /* current pattern segment large enough to contain full srcPatternLength */
+ && (forwardPatternLength <= srcPatternLength) ) { /* haven't reached this position yet */
+#if 1
+ matchIndex += (U32)forwardPatternLength - (U32)srcPatternLength; /* best position, full pattern, might be followed by more match */
+#else
+ const BYTE* const matchCandidate = matchPtr + (U32)forwardPatternLength - (U32)srcPatternLength; /* best position, pattern might be followed by more match */
+ int matchLength = (int)(LZ4_count(ip + srcPatternLength, matchCandidate + srcPatternLength, iHighLimit) + srcPatternLength);
+ int back = 0;
+ while ( (ip+back > iLowLimit)
+ && (matchPtr+back > lowPrefixPtr)
+ && (ip[back-1] == matchPtr[back-1])) {
+ back--;
+ }
+ matchLength -= back;
+ if (matchLength > longest) {
+ longest = matchLength;
+ *matchpos = base + matchIndex + back;
+ *startpos = ip + back;
+ }
+ matchIndex -= (U32)backLength;
+ matchIndex -= DELTANEXTU16(chainTable, matchIndex); /* skip directly to next potential pattern segment */
+#endif
+ } else {
+ matchIndex -= (U32)backLength; /* let's go to farthest segment position, will find a match of length currentSegmentLength + maybe some back */
+ //matchIndex -= DELTANEXTU16(chainTable, matchIndex); /* skip directly to following candidate; slightly faster, but miss some rare corner cases (likely when back is useful)*/
+ }
+ } } } }
+ } /* while ((matchIndex>=lowLimit) && (nbAttempts)) */
return longest;
}