Add ZSTD_d_maxBlockSize parameter

Reduces memory when blocks are guaranteed to be smaller than allowed by
the format. This is useful for streaming compression in conjunction with
ZSTD_c_maxBlockSize.

This PR saves 2 * (formatMaxBlockSize - paramMaxBlockSize) when streaming.
Once it is rebased on top of PR #3616 it will save
3 * (formatMaxBlockSize - paramMaxBlockSize).

diff --git a/lib/decompress/zstd_decompress.c b/lib/decompress/zstd_decompress.c
index 94b90481b525ad7ccaab37a14db6ad0ae1f9d456..94eb95151b53da4b421a8cb62b64f5280a72c19d 100644 (file)
--- a/lib/decompress/zstd_decompress.c
+++ b/lib/decompress/zstd_decompress.c
@@ -245,6 +245,7 @@ static void ZSTD_DCtx_resetParameters(ZSTD_DCtx* dctx)
     dctx->forceIgnoreChecksum = ZSTD_d_validateChecksum;
     dctx->refMultipleDDicts = ZSTD_rmd_refSingleDDict;
     dctx->disableHufAsm = 0;
+    dctx->maxBlockSizeParam = 0;
 }
 
 static void ZSTD_initDCtx_internal(ZSTD_DCtx* dctx)
@@ -972,6 +973,10 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx,
         ip += frameHeaderSize; remainingSrcSize -= frameHeaderSize;
     }
 
+    /* Shrink the blockSizeMax if enabled */
+    if (dctx->maxBlockSizeParam != 0)
+        dctx->fParams.blockSizeMax = MIN(dctx->fParams.blockSizeMax, (unsigned)dctx->maxBlockSizeParam);
+
     /* Loop on each block */
     while (1) {
         BYTE* oBlockEnd = oend;
@@ -1823,6 +1828,10 @@ ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam)
             bounds.lowerBound = 0;
             bounds.upperBound = 1;
             return bounds;
+        case ZSTD_d_maxBlockSize:
+            bounds.lowerBound = ZSTD_BLOCKSIZE_MAX_MIN;
+            bounds.upperBound = ZSTD_BLOCKSIZE_MAX;
+            return bounds;
 
         default:;
     }
@@ -1867,6 +1876,9 @@ size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value
         case ZSTD_d_disableHuffmanAssembly:
             *value = (int)dctx->disableHufAsm;
             return 0;
+        case ZSTD_d_maxBlockSize:
+            *value = dctx->maxBlockSizeParam;
+            return 0;
         default:;
     }
     RETURN_ERROR(parameter_unsupported, "");
@@ -1904,6 +1916,10 @@ size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter dParam, int value
             CHECK_DBOUNDS(ZSTD_d_disableHuffmanAssembly, value);
             dctx->disableHufAsm = value != 0;
             return 0;
+        case ZSTD_d_maxBlockSize:
+            if (value != 0) CHECK_DBOUNDS(ZSTD_d_maxBlockSize, value);
+            dctx->maxBlockSizeParam = value;
+            return 0;
         default:;
     }
     RETURN_ERROR(parameter_unsupported, "");
@@ -1932,9 +1948,9 @@ size_t ZSTD_sizeof_DStream(const ZSTD_DStream* dctx)
     return ZSTD_sizeof_DCtx(dctx);
 }
 
-size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize)
+static size_t ZSTD_decodingBufferSize_internal(unsigned long long windowSize, unsigned long long frameContentSize, size_t blockSizeMax)
 {
-    size_t const blockSize = (size_t)MIN(windowSize, ZSTD_BLOCKSIZE_MAX);
+    size_t const blockSize = MIN((size_t)MIN(windowSize, ZSTD_BLOCKSIZE_MAX), blockSizeMax);
     /* We need blockSize + WILDCOPY_OVERLENGTH worth of buffer so that if a block
      * ends at windowSize + WILDCOPY_OVERLENGTH + 1 bytes, we can start writing
      * the block at the beginning of the output buffer, and maintain a full window.
@@ -1950,6 +1966,11 @@ size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long
     return minRBSize;
 }
 
+size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize)
+{
+    return ZSTD_decodingBufferSize_internal(windowSize, frameContentSize, ZSTD_BLOCKSIZE_MAX);
+}
+
 size_t ZSTD_estimateDStreamSize(size_t windowSize)
 {
     size_t const blockSize = MIN(windowSize, ZSTD_BLOCKSIZE_MAX);
@@ -2188,11 +2209,13 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB
             zds->fParams.windowSize = MAX(zds->fParams.windowSize, 1U << ZSTD_WINDOWLOG_ABSOLUTEMIN);
             RETURN_ERROR_IF(zds->fParams.windowSize > zds->maxWindowSize,
                             frameParameter_windowTooLarge, "");
+            if (zds->maxBlockSizeParam != 0)
+                zds->fParams.blockSizeMax = MIN(zds->fParams.blockSizeMax, (unsigned)zds->maxBlockSizeParam);
 
             /* Adapt buffer sizes to frame header instructions */
             {   size_t const neededInBuffSize = MAX(zds->fParams.blockSizeMax, 4 /* frame checksum */);
                 size_t const neededOutBuffSize = zds->outBufferMode == ZSTD_bm_buffered
-                        ? ZSTD_decodingBufferSize_min(zds->fParams.windowSize, zds->fParams.frameContentSize)
+                        ? ZSTD_decodingBufferSize_internal(zds->fParams.windowSize, zds->fParams.frameContentSize, zds->fParams.blockSizeMax)
                         : 0;
 
                 ZSTD_DCtx_updateOversizedDuration(zds, neededInBuffSize, neededOutBuffSize);

diff --git a/lib/decompress/zstd_decompress_internal.h b/lib/decompress/zstd_decompress_internal.h
index 25aaebe500180d2f57b4bbd7bd63eab0287c01f3..83a7a0115fdba0fbc26b9a4933af292734ac8e62 100644 (file)
--- a/lib/decompress/zstd_decompress_internal.h
+++ b/lib/decompress/zstd_decompress_internal.h
@@ -167,6 +167,7 @@ struct ZSTD_DCtx_s
     ZSTD_DDictHashSet* ddictSet;                    /* Hash set for multiple ddicts */
     ZSTD_refMultipleDDicts_e refMultipleDDicts;     /* User specified: if == 1, will allow references to multiple DDicts. Default == 0 (disabled) */
     int disableHufAsm;
+    int maxBlockSizeParam;
 
     /* streaming */
     ZSTD_dStreamStage streamStage;

diff --git a/lib/zstd.h b/lib/zstd.h
index e5c3f8b68b72e926b0720d7a2efec3f94792d92e..e94f101897051ac90495bafdc206e7ba1971a6cf 100644 (file)
--- a/lib/zstd.h
+++ b/lib/zstd.h
@@ -618,6 +618,7 @@ typedef enum {
      * ZSTD_d_forceIgnoreChecksum
      * ZSTD_d_refMultipleDDicts
      * ZSTD_d_disableHuffmanAssembly
+     * ZSTD_d_maxBlockSize
      * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them.
      * note : never ever use experimentalParam? names directly
      */
@@ -625,7 +626,8 @@ typedef enum {
      ZSTD_d_experimentalParam2=1001,
      ZSTD_d_experimentalParam3=1002,
      ZSTD_d_experimentalParam4=1003,
-     ZSTD_d_experimentalParam5=1004
+     ZSTD_d_experimentalParam5=1004,
+     ZSTD_d_experimentalParam6=1005
 
 } ZSTD_dParameter;
 
@@ -2430,6 +2432,22 @@ ZSTDLIB_STATIC_API size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParamete
  */
 #define ZSTD_d_disableHuffmanAssembly ZSTD_d_experimentalParam5
 
+/* ZSTD_d_maxBlockSize
+ * Allowed values are between 1KB and ZSTD_BLOCKSIZE_MAX (128KB).
+ * The default is ZSTD_BLOCKSIZE_MAX, and setting to 0 will set to the default.
+ *
+ * Forces the decompressor to reject blocks whose content size is
+ * larger than the configured maxBlockSize. When maxBlockSize is
+ * larger than the windowSize, the windowSize is used instead.
+ * This saves memory on the decoder when you know all blocks are small.
+ *
+ * This option is typically used in conjunction with ZSTD_c_maxBlockSize.
+ *
+ * WARNING: This causes the decoder to reject otherwise valid frames
+ * that have block sizes larger than the configured maxBlockSize.
+ */
+#define ZSTD_d_maxBlockSize ZSTD_d_experimentalParam6
+
 
 /*! ZSTD_DCtx_setFormat() :
  *  This function is REDUNDANT. Prefer ZSTD_DCtx_setParameter().

diff --git a/tests/fuzz/simple_round_trip.c b/tests/fuzz/simple_round_trip.c
index 0fe46391764411a3bcd44ab3a7db354ab0f9b1af..660092e6106c8c2a0d736723f63adb87dbda4cab 100644 (file)
--- a/tests/fuzz/simple_round_trip.c
+++ b/tests/fuzz/simple_round_trip.c
@@ -90,6 +90,9 @@ static size_t roundTripTest(void *result, size_t resultCapacity,
             FUZZ_ASSERT(XXH64(compressed, cSize, 0) == hash0);
         }
     }
+    if (FUZZ_dataProducer_uint32Range(producer, 0, 1)) {
+        FUZZ_ZASSERT(ZSTD_DCtx_setParameter(dctx, ZSTD_d_maxBlockSize, maxBlockSize));
+    }
     dSize = ZSTD_decompressDCtx(dctx, result, resultCapacity, compressed, cSize);
     FUZZ_ZASSERT(dSize);
     FUZZ_ASSERT_MSG(dSize == srcSize, "Incorrect regenerated size");

diff --git a/tests/fuzz/stream_round_trip.c b/tests/fuzz/stream_round_trip.c
index 7d277a857a671334a8ca0ceb254df5a7c240ed97..c2d6707a1ced37592838d6e3cf442b4ed107344b 100644 (file)
--- a/tests/fuzz/stream_round_trip.c
+++ b/tests/fuzz/stream_round_trip.c
@@ -63,6 +63,8 @@ static size_t compress(uint8_t *dst, size_t capacity,
     size_t dstSize = 0;
     ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only);
     FUZZ_setRandomParameters(cctx, srcSize, producer);
+    int maxBlockSize;
+    FUZZ_ZASSERT(ZSTD_CCtx_getParameter(cctx, ZSTD_c_maxBlockSize, &maxBlockSize));
 
     while (srcSize > 0) {
         ZSTD_inBuffer in = makeInBuffer(&src, &srcSize, producer);
@@ -93,6 +95,8 @@ static size_t compress(uint8_t *dst, size_t capacity,
                         if (FUZZ_dataProducer_uint32Range(producer, 0, 7) == 0) {
                             size_t const remaining = in.size - in.pos;
                             FUZZ_setRandomParameters(cctx, remaining, producer);
+                            /* Always use the same maxBlockSize */
+                            FUZZ_ZASSERT(ZSTD_CCtx_setParameter(cctx, ZSTD_c_maxBlockSize, maxBlockSize));
                         }
                         mode = -1;
                     }
@@ -132,6 +136,23 @@ static size_t compress(uint8_t *dst, size_t capacity,
     return dstSize;
 }
 
+size_t decompress(void* dst, size_t dstCapacity, void const* src, size_t srcSize, FUZZ_dataProducer_t* producer)
+{
+    ZSTD_inBuffer in = {src, srcSize, 0};
+    ZSTD_outBuffer out = {dst, dstCapacity, 0};
+    int maxBlockSize;
+    FUZZ_ZASSERT(ZSTD_CCtx_getParameter(cctx, ZSTD_c_maxBlockSize, &maxBlockSize));
+    if (FUZZ_dataProducer_uint32Range(producer, 0, 1)) {
+        FUZZ_ZASSERT(ZSTD_DCtx_setParameter(dctx, ZSTD_d_maxBlockSize, maxBlockSize));
+    }
+    while (in.pos < in.size) {
+        size_t const ret = ZSTD_decompressStream(dctx, &out, &in);
+        FUZZ_ZASSERT(ret);
+        FUZZ_ASSERT(ret == 0);
+    }
+    return out.pos;
+}
+
 int LLVMFuzzerTestOneInput(const uint8_t *src, size_t size)
 {
     FUZZ_SEQ_PROD_SETUP();
@@ -163,8 +184,7 @@ int LLVMFuzzerTestOneInput(const uint8_t *src, size_t size)
 
     {
         size_t const cSize = compress(cBuf, neededBufSize, src, size, producer);
-        size_t const rSize =
-            ZSTD_decompressDCtx(dctx, rBuf, neededBufSize, cBuf, cSize);
+        size_t const rSize = decompress(rBuf, neededBufSize, cBuf, cSize, producer);
         FUZZ_ZASSERT(rSize);
         FUZZ_ASSERT_MSG(rSize == size, "Incorrect regenerated size");
         FUZZ_ASSERT_MSG(!FUZZ_memcmp(src, rBuf, size), "Corruption!");

diff --git a/tests/fuzzer.c b/tests/fuzzer.c
index 07ddfefd6db9be5f450148765b0a99a7d4e9ddb3..b049a9e09ba2590c5c4888afa128f82e6d8184ae 100644 (file)
--- a/tests/fuzzer.c
+++ b/tests/fuzzer.c
@@ -952,6 +952,25 @@ static int basicUnitTests(U32 const seed, double compressibility)
         ZSTD_freeCDict(cdict);
         ZSTD_freeCCtx(cctx);
     }
+    
+    DISPLAYLEVEL(3, "test%3i : maxBlockSize = 2K", testNb++);
+    {
+        ZSTD_CCtx* cctx = ZSTD_createCCtx();
+        ZSTD_DCtx* dctx = ZSTD_createDCtx();
+        CHECK_Z(ZSTD_CCtx_setParameter(cctx, ZSTD_c_checksumFlag, 1));
+        CHECK_Z(ZSTD_CCtx_setParameter(cctx, ZSTD_c_maxBlockSize, 2048));
+        CHECK_Z(ZSTD_DCtx_setParameter(dctx, ZSTD_d_maxBlockSize, 2048));
+
+        cSize = ZSTD_compress2(cctx, compressedBuffer, compressedBufferSize, CNBuffer, CNBuffSize);
+        CHECK_Z(cSize);
+        CHECK_Z(ZSTD_decompressDCtx(dctx, decodedBuffer, CNBuffSize, compressedBuffer, cSize));
+
+        CHECK_Z(ZSTD_DCtx_setParameter(dctx, ZSTD_d_maxBlockSize, 1024));
+        CHECK(ZSTD_isError(ZSTD_decompressDCtx(dctx, decodedBuffer, CNBuffSize, compressedBuffer, cSize)));
+
+        ZSTD_freeDCtx(dctx);
+        ZSTD_freeCCtx(cctx);
+    }
 
     DISPLAYLEVEL(3, "test%3i : ldm fill dict out-of-bounds check", testNb++);
     {

diff --git a/tests/zstreamtest.c b/tests/zstreamtest.c
index 14c4af82fb76c1fb5675829668cc70a3e17b72ed..85d0fc81fe0efc2af0647665f77563f16d0fb9cd 100644 (file)
--- a/tests/zstreamtest.c
+++ b/tests/zstreamtest.c
@@ -722,6 +722,67 @@ static int basicUnitTests(U32 seed, double compressibility, int bigTests)
     }
     DISPLAYLEVEL(3, "OK \n");
 
+    DISPLAYLEVEL(3, "test%3i : maxBlockSize = 2KB : ", testNb++);
+    {
+        ZSTD_DCtx* dctx = ZSTD_createDCtx();
+        size_t singlePassSize, streamingSize, streaming2KSize;
+
+        {
+            ZSTD_CCtx* cctx = ZSTD_createCCtx();
+            CHECK_Z(ZSTD_CCtx_setParameter(cctx, ZSTD_c_checksumFlag, 1));
+            CHECK_Z(ZSTD_CCtx_setParameter(cctx, ZSTD_c_windowLog, 18));
+            CHECK_Z(ZSTD_CCtx_setParameter(cctx, ZSTD_c_contentSizeFlag, 0));
+            CHECK_Z(ZSTD_CCtx_setParameter(cctx, ZSTD_c_maxBlockSize, 2048));
+            cSize = ZSTD_compress2(cctx, compressedBuffer, compressedBufferSize, CNBuffer, CNBufferSize);
+            CHECK_Z(cSize);
+            ZSTD_freeCCtx(cctx);
+        }
+
+        CHECK_Z(ZSTD_decompressDCtx(dctx, decodedBuffer, CNBufferSize, compressedBuffer, cSize));
+        singlePassSize = ZSTD_sizeof_DCtx(dctx);
+        CHECK_Z(singlePassSize);
+
+        inBuff.src = compressedBuffer;
+        inBuff.size = cSize;
+
+        outBuff.dst = decodedBuffer;
+        outBuff.size = decodedBufferSize;
+
+        CHECK_Z(ZSTD_DCtx_setParameter(dctx, ZSTD_d_maxBlockSize, 2048));
+        inBuff.pos = 0;
+        outBuff.pos = 0;
+        {
+            size_t const r = ZSTD_decompressStream(dctx, &outBuff, &inBuff);
+            CHECK_Z(r);
+            CHECK(r != 0, "Entire frame must be decompressed");
+        }
+        streaming2KSize = ZSTD_sizeof_DCtx(dctx);
+        CHECK_Z(streaming2KSize);
+        
+        CHECK_Z(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_and_parameters));
+        inBuff.pos = 0;
+        outBuff.pos = 0;
+        {
+            size_t const r = ZSTD_decompressStream(dctx, &outBuff, &inBuff);
+            CHECK_Z(r);
+            CHECK(r != 0, "Entire frame must be decompressed");
+        }
+        streamingSize = ZSTD_sizeof_DCtx(dctx);
+        CHECK_Z(streamingSize);
+        
+        CHECK_Z(ZSTD_DCtx_setParameter(dctx, ZSTD_d_maxBlockSize, 1024));
+        inBuff.pos = 0;
+        outBuff.pos = 0;
+        CHECK(!ZSTD_isError(ZSTD_decompressStream(dctx, &outBuff, &inBuff)), "decompression must fail");
+
+        CHECK(streamingSize < singlePassSize + (1 << 18) + 3 * ZSTD_BLOCKSIZE_MAX, "Streaming doesn't use the right amount of memory");
+        CHECK(streamingSize != streaming2KSize + 3 * (ZSTD_BLOCKSIZE_MAX - 2048), "ZSTD_d_blockSizeMax didn't save the right amount of memory");
+        DISPLAYLEVEL(3, "| %zu | %zu | %zu | ", singlePassSize, streaming2KSize, streamingSize);
+
+        ZSTD_freeDCtx(dctx);
+    }
+    DISPLAYLEVEL(3, "OK \n");
+
     /* Decompression with ZSTD_d_stableOutBuffer */
     cSize = ZSTD_compress(compressedBuffer, compressedBufferSize, CNBuffer, CNBufferSize, 1);
     CHECK_Z(cSize);
@@ -2845,6 +2906,13 @@ static int fuzzerTests_newAPI(U32 seed, int nbTests, int startTest,
                 if (FUZ_rand(&lseed) & 1) CHECK_Z( setCCtxParameter(zc, cctxParams, ZSTD_c_forceMaxWindow, FUZ_rand(&lseed) & 1, opaqueAPI) );
                 if (FUZ_rand(&lseed) & 1) CHECK_Z( setCCtxParameter(zc, cctxParams, ZSTD_c_deterministicRefPrefix, FUZ_rand(&lseed) & 1, opaqueAPI) );
 
+                /* Set max block size parameters */
+                if (FUZ_rand(&lseed) & 1) {
+                    int maxBlockSize = (int)(FUZ_rand(&lseed) % ZSTD_BLOCKSIZE_MAX);
+                    maxBlockSize = MAX(1024, maxBlockSize);
+                    CHECK_Z( setCCtxParameter(zc, cctxParams, ZSTD_c_maxBlockSize, maxBlockSize, opaqueAPI) );
+                }
+
                 /* Apply parameters */
                 if (opaqueAPI) {
                     DISPLAYLEVEL(5, "t%u: applying CCtxParams \n", testNb);
@@ -2976,6 +3044,13 @@ static int fuzzerTests_newAPI(U32 seed, int nbTests, int startTest,
         if (FUZ_rand(&lseed) & 1) {
             CHECK_Z(ZSTD_DCtx_setParameter(zd, ZSTD_d_disableHuffmanAssembly, FUZ_rand(&lseed) & 1));
         }
+        if (FUZ_rand(&lseed) & 1) {
+            int maxBlockSize;
+            CHECK_Z(ZSTD_CCtx_getParameter(zc, ZSTD_c_maxBlockSize, &maxBlockSize));
+            CHECK_Z(ZSTD_DCtx_setParameter(zd, ZSTD_d_maxBlockSize, maxBlockSize));
+        } else {
+            CHECK_Z(ZSTD_DCtx_setParameter(zd, ZSTD_d_maxBlockSize, 0));
+        }
         {   size_t decompressionResult = 1;
             ZSTD_inBuffer  inBuff = { cBuffer, cSize, 0 };
             ZSTD_outBuffer outBuff= { dstBuffer, dstBufferSize, 0 };