[lib] Fix off-by-one error in repcode checks

The repcode checks disallowed repcodes that are equal to `windowLow`.
This is slightly inefficient, but isn't a problem on its own. Together
with the next commit, it cause non-determinism.

diff --git a/lib/compress/zstd_double_fast.c b/lib/compress/zstd_double_fast.c
index d0d3a784dd2fec7b8192dd4ceb34f8fcb9981d4b..b9e3c5e9f4f80beda233ff0a9acdc89817a2abab 100644 (file)
--- a/lib/compress/zstd_double_fast.c
+++ b/lib/compress/zstd_double_fast.c
@@ -409,7 +409,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
         hashSmall[hSmall] = hashLong[hLong] = curr;   /* update hash table */
 
         if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */
-            & (offset_1 < curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */
+            & (offset_1 <= curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */
           && (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;
@@ -477,7 +477,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                 U32 const repIndex2 = current2 - offset_2;
                 const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
                 if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3)   /* intentional overflow : ensure repIndex2 doesn't overlap dict + prefix */
-                    & (offset_2 < current2 - dictStartIndex))
+                    & (offset_2 <= current2 - dictStartIndex))
                   && (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;

diff --git a/lib/compress/zstd_fast.c b/lib/compress/zstd_fast.c
index 4edc04dccdfbcdd9b676d52f439f609f6ac32166..b6a1a7e696b1f70b86d1d5e6070273c0e07622cd 100644 (file)
--- a/lib/compress/zstd_fast.c
+++ b/lib/compress/zstd_fast.c
@@ -418,7 +418,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
         DEBUGLOG(7, "offset_1 = %u , curr = %u", offset_1, curr);
 
         if ( ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow */
-             & (offset_1 < curr+1 - dictStartIndex) ) /* note: we are searching at curr+1 */
+             & (offset_1 <= curr+1 - dictStartIndex) ) /* note: we are searching at curr+1 */
            && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
             const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
             size_t const rLength = ZSTD_count_2segments(ip+1 +4, repMatch +4, iend, repMatchEnd, prefixStart) + 4;
@@ -453,7 +453,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
                 U32 const current2 = (U32)(ip-base);
                 U32 const repIndex2 = current2 - offset_2;
                 const BYTE* const repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
-                if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (offset_2 < curr - dictStartIndex))  /* intentional overflow */
+                if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (offset_2 <= curr - 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;

diff --git a/lib/compress/zstd_lazy.c b/lib/compress/zstd_lazy.c
index 3d523e8472f68aba16009d91a10ff06bde0daaf2..0c4b49e4da0ec3bd10bd60af7fe4238dc6758a03 100644 (file)
--- a/lib/compress/zstd_lazy.c
+++ b/lib/compress/zstd_lazy.c
@@ -1995,7 +1995,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
             const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
             const BYTE* const repMatch = repBase + repIndex;
             if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow */
-               & (offset_1 < curr+1 - windowLow) ) /* note: we are searching at curr+1 */
+               & (offset_1 <= curr+1 - windowLow) ) /* note: we are searching at curr+1 */
             if (MEM_read32(ip+1) == MEM_read32(repMatch)) {
                 /* repcode detected we should take it */
                 const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
@@ -2027,7 +2027,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
                 const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
                 const BYTE* const repMatch = repBase + repIndex;
                 if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments  */
-                   & (offset_1 < curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */
+                   & (offset_1 <= curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */
                 if (MEM_read32(ip) == MEM_read32(repMatch)) {
                     /* repcode detected */
                     const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
@@ -2059,7 +2059,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
                     const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
                     const BYTE* const repMatch = repBase + repIndex;
                     if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments  */
-                       & (offset_1 < curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */
+                       & (offset_1 <= curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */
                     if (MEM_read32(ip) == MEM_read32(repMatch)) {
                         /* repcode detected */
                         const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
@@ -2106,7 +2106,7 @@ _storeSequence:
             const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
             const BYTE* const repMatch = repBase + repIndex;
             if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments  */
-               & (offset_2 < repCurrent - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */
+               & (offset_2 <= repCurrent - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */
             if (MEM_read32(ip) == MEM_read32(repMatch)) {
                 /* repcode detected we should take it */
                 const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;