Adding --long support for --patch-from

lib/compress/zstd_compress.c

@@ -945,9 +945,11 @@ size_t ZSTD_CCtx_refPrefix_advanced(
 {
     RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
     ZSTD_clearAllDicts(cctx);
-    cctx->prefixDict.dict = prefix;
-    cctx->prefixDict.dictSize = prefixSize;
-    cctx->prefixDict.dictContentType = dictContentType;
+    if (prefix != NULL && prefixSize > 0) {
+        cctx->prefixDict.dict = prefix;
+        cctx->prefixDict.dictSize = prefixSize;
+        cctx->prefixDict.dictContentType = dictContentType;
+    }
     return 0;
 }
 
@@ -2782,6 +2784,7 @@ size_t ZSTD_compressBlock(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const
  *  @return : 0, or an error code
  */
 static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms,
+                                         ldmState_t* ls,
                                          ZSTD_cwksp* ws,
                                          ZSTD_CCtx_params const* params,
                                          const void* src, size_t srcSize,
@@ -2793,6 +2796,11 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms,
     ZSTD_window_update(&ms->window, src, srcSize);
     ms->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ms->window.base);
 
+    if (params->ldmParams.enableLdm && ls != NULL) {
+        ZSTD_window_update(&ls->window, src, srcSize);
+        ls->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ls->window.base);
+    }
+
     /* Assert that we the ms params match the params we're being given */
     ZSTD_assertEqualCParams(params->cParams, ms->cParams);
 
@@ -2805,6 +2813,11 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms,
 
         ZSTD_overflowCorrectIfNeeded(ms, ws, params, ip, ichunk);
 
+        if (params->ldmParams.enableLdm && ls != NULL && srcSize >= params->ldmParams.minMatchLength) {
+            ZSTD_ldm_fillHashTable(ls, (const BYTE*)src, (const BYTE*)src + srcSize, &params->ldmParams);
+            break;
+        }
+
         switch(params->cParams.strategy)
         {
         case ZSTD_fast:
@@ -2983,7 +2996,7 @@ static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs,
         bs->entropy.fse.matchlength_repeatMode = FSE_repeat_valid;
         bs->entropy.fse.litlength_repeatMode = FSE_repeat_valid;
         FORWARD_IF_ERROR(ZSTD_loadDictionaryContent(
-            ms, ws, params, dictPtr, dictContentSize, dtlm));
+            ms, NULL, ws, params, dictPtr, dictContentSize, dtlm));
         return dictID;
     }
 }
@@ -2993,6 +3006,7 @@ static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs,
 static size_t
 ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs,
                                ZSTD_matchState_t* ms,
+                               ldmState_t* ls,
                                ZSTD_cwksp* ws,
                          const ZSTD_CCtx_params* params,
                          const void* dict, size_t dictSize,
@@ -3010,13 +3024,13 @@ ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs,
 
     /* dict restricted modes */
     if (dictContentType == ZSTD_dct_rawContent)
-        return ZSTD_loadDictionaryContent(ms, ws, params, dict, dictSize, dtlm);
+        return ZSTD_loadDictionaryContent(ms, ls, ws, params, dict, dictSize, dtlm);
 
     if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) {
         if (dictContentType == ZSTD_dct_auto) {
             DEBUGLOG(4, "raw content dictionary detected");
             return ZSTD_loadDictionaryContent(
-                ms, ws, params, dict, dictSize, dtlm);
+                ms, ls, ws, params, dict, dictSize, dtlm);
         }
         RETURN_ERROR_IF(dictContentType == ZSTD_dct_fullDict, dictionary_wrong);
         assert(0);   /* impossible */
@@ -3059,12 +3073,12 @@ static size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx,
     {   size_t const dictID = cdict ?
                 ZSTD_compress_insertDictionary(
                         cctx->blockState.prevCBlock, &cctx->blockState.matchState,
-                        &cctx->workspace, &cctx->appliedParams, cdict->dictContent,
+                        &cctx->ldmState, &cctx->workspace, &cctx->appliedParams, cdict->dictContent,
                         cdict->dictContentSize, dictContentType, dtlm,
                         cctx->entropyWorkspace)
               : ZSTD_compress_insertDictionary(
                         cctx->blockState.prevCBlock, &cctx->blockState.matchState,
-                        &cctx->workspace, &cctx->appliedParams, dict, dictSize,
+                        &cctx->ldmState, &cctx->workspace, &cctx->appliedParams, dict, dictSize,
                         dictContentType, dtlm, cctx->entropyWorkspace);
         FORWARD_IF_ERROR(dictID);
         assert(dictID <= UINT_MAX);
@@ -3342,7 +3356,7 @@ static size_t ZSTD_initCDict_internal(
         params.fParams.contentSizeFlag = 1;
         params.cParams = cParams;
         {   size_t const dictID = ZSTD_compress_insertDictionary(
-                    &cdict->cBlockState, &cdict->matchState, &cdict->workspace,
+                    &cdict->cBlockState, &cdict->matchState, NULL, &cdict->workspace,
                     &params, cdict->dictContent, cdict->dictContentSize,
                     dictContentType, ZSTD_dtlm_full, cdict->entropyWorkspace);
             FORWARD_IF_ERROR(dictID);

lib/compress/zstd_compress_internal.h

@@ -166,6 +166,7 @@ typedef struct {
 typedef struct {
     ZSTD_window_t window;   /* State for the window round buffer management */
     ldmEntry_t* hashTable;
+    U32 loadedDictEnd;
     BYTE* bucketOffsets;    /* Next position in bucket to insert entry */
     U64 hashPower;          /* Used to compute the rolling hash.
                              * Depends on ldmParams.minMatchLength */

lib/compress/zstd_ldm.c

@@ -224,6 +224,17 @@ static U64 ZSTD_ldm_fillLdmHashTable(ldmState_t* state,
     return rollingHash;
 }
 
+void ZSTD_ldm_fillHashTable(
+            ldmState_t* state, const BYTE* ip,
+            const BYTE* iend, ldmParams_t const* params)
+{
+    U64 startingHash = ZSTD_rollingHash_compute(ip, params->minMatchLength);
+    ZSTD_ldm_fillLdmHashTable(
+        state, startingHash, ip, iend - params->minMatchLength, state->window.base,
+        params->hashLog - params->bucketSizeLog,
+        *params);
+}
+
 
 /** ZSTD_ldm_limitTableUpdate() :
  *
@@ -459,7 +470,7 @@ size_t ZSTD_ldm_generateSequences(
          *       * Try invalidation after the sequence generation and test the
          *         the offset against maxDist directly.
          */
-        ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, NULL, NULL);
+        ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, &ldmState->loadedDictEnd, NULL);
         /* 3. Generate the sequences for the chunk, and get newLeftoverSize. */
         newLeftoverSize = ZSTD_ldm_generateSequences_internal(
             ldmState, sequences, params, chunkStart, chunkSize);
@@ -567,7 +578,6 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
         if (sequence.offset == 0)
             break;
 
-        assert(sequence.offset <= (1U << cParams->windowLog));
         assert(ip + sequence.litLength + sequence.matchLength <= iend);
 
         /* Fill tables for block compressor */

lib/compress/zstd_ldm.h

@@ -24,6 +24,10 @@ extern "C" {
 
 #define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
 
+void ZSTD_ldm_fillHashTable(
+            ldmState_t* state, const BYTE* ip,
+            const BYTE* iend, ldmParams_t const* params);
+
 /**
  * ZSTD_ldm_generateSequences():
  *

lib/compress/zstdmt_compress.c

@@ -461,7 +461,9 @@ typedef struct {
     ZSTD_window_t ldmWindow;  /* A thread-safe copy of ldmState.window */
 } serialState_t;
 
-static int ZSTDMT_serialState_reset(serialState_t* serialState, ZSTDMT_seqPool* seqPool, ZSTD_CCtx_params params, size_t jobSize)
+static int ZSTDMT_serialState_reset(serialState_t* serialState, 
+                        ZSTDMT_seqPool* seqPool, ZSTD_CCtx_params params, 
+                        size_t jobSize, const void* dict, size_t const dictSize)
 {
     /* Adjust parameters */
     if (params.ldmParams.enableLdm) {
@@ -507,6 +509,13 @@ static int ZSTDMT_serialState_reset(serialState_t* serialState, ZSTDMT_seqPool*
         memset(serialState->ldmState.hashTable, 0, hashSize);
         memset(serialState->ldmState.bucketOffsets, 0, bucketSize);
     }
+
+    /* Update window state and fill hash table with dict */
+    if (params.ldmParams.enableLdm && dict) {
+        ZSTD_window_update(&serialState->ldmState.window, dict, dictSize);
+        ZSTD_ldm_fillHashTable(&serialState->ldmState, (const BYTE*)dict, (const BYTE*)dict + dictSize, &params.ldmParams);
+    }
+
     serialState->params = params;
     serialState->params.jobSize = (U32)jobSize;
     return 0;
@@ -1267,7 +1276,7 @@ static size_t ZSTDMT_compress_advanced_internal(
 
     assert(avgJobSize >= 256 KB);  /* condition for ZSTD_compressBound(A) + ZSTD_compressBound(B) <= ZSTD_compressBound(A+B), required to compress directly into Dst (no additional buffer) */
     ZSTDMT_setBufferSize(mtctx->bufPool, ZSTD_compressBound(avgJobSize) );
-    if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, avgJobSize))
+    if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, avgJobSize, NULL, 0))
         return ERROR(memory_allocation);
 
     FORWARD_IF_ERROR( ZSTDMT_expandJobsTable(mtctx, nbJobs) );  /* only expands if necessary */
@@ -1500,7 +1509,7 @@ size_t ZSTDMT_initCStream_internal(
     mtctx->allJobsCompleted = 0;
     mtctx->consumed = 0;
     mtctx->produced = 0;
-    if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, mtctx->targetSectionSize))
+    if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, mtctx->targetSectionSize, dict, dictSize))
         return ERROR(memory_allocation);
     return 0;
 }