Benchmark program for sequence compression API

diff --git a/contrib/seqBench/Makefile b/contrib/seqBench/Makefile
new file mode 100644 (file)
index 0000000..0782961
--- /dev/null
+++ b/contrib/seqBench/Makefile
@@ -0,0 +1,58 @@
+# ################################################################
+# Copyright (c) 2018-present, Yann Collet, Facebook, Inc.
+# All rights reserved.
+#
+# This source code is licensed under both the BSD-style license (found in the
+# LICENSE file in the root directory of this source tree) and the GPLv2 (found
+# in the COPYING file in the root directory of this source tree).
+# ################################################################
+
+PROGDIR = ../../programs
+LIBDIR  = ../../lib
+
+LIBZSTD = $(LIBDIR)/libzstd.a
+
+CPPFLAGS+= -I$(LIBDIR) -I$(LIBDIR)/common -I$(LIBDIR)/dictBuilder -I$(PROGDIR)
+
+CFLAGS  ?= -O3 -g
+CFLAGS  += -std=gnu99
+DEBUGFLAGS= -Wall -Wextra -Wcast-qual -Wcast-align -Wshadow \
+            -Wstrict-aliasing=1 -Wswitch-enum \
+            -Wstrict-prototypes -Wundef -Wpointer-arith \
+            -Wvla -Wformat=2 -Winit-self -Wfloat-equal -Wwrite-strings \
+            -Wredundant-decls
+CFLAGS  += $(DEBUGFLAGS) $(MOREFLAGS)
+
+
+default: seqBench
+
+all : seqBench
+
+seqBench: util.o timefn.o benchfn.o datagen.o xxhash.o seqBench.c $(LIBZSTD)
+       $(CC) $(CPPFLAGS) $(CFLAGS) $^ $(LDFLAGS) -o $@
+
+.PHONY: $(LIBZSTD)
+$(LIBZSTD):
+       $(MAKE) -C $(LIBDIR) libzstd.a CFLAGS="$(CFLAGS)"
+
+benchfn.o: $(PROGDIR)/benchfn.c
+       $(CC) $(CPPFLAGS) $(CFLAGS) $^ -c
+
+timefn.o: $(PROGDIR)/timefn.c
+       $(CC) $(CPPFLAGS) $(CFLAGS) $^ -c
+
+datagen.o: $(PROGDIR)/datagen.c
+       $(CC) $(CPPFLAGS) $(CFLAGS) $^ -c
+
+util.o: $(PROGDIR)/util.c
+       $(CC) $(CPPFLAGS) $(CFLAGS) $^ -c
+
+
+xxhash.o : $(LIBDIR)/common/xxhash.c
+       $(CC) $(CPPFLAGS) $(CFLAGS) $^ -c
+
+
+clean:
+       $(RM) *.o
+       $(MAKE) -C $(LIBDIR) clean > /dev/null
+       $(RM) seqBench

diff --git a/contrib/seqBench/seqBench.c b/contrib/seqBench/seqBench.c
new file mode 100644 (file)
index 0000000..473649c
--- /dev/null
+++ b/contrib/seqBench/seqBench.c
@@ -0,0 +1,55 @@
+#define ZSTD_STATIC_LINKING_ONLY
+#include <zstd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <assert.h>
+#include <string.h>
+
+int main(int argc, char *argv[]) {
+    ZSTD_CCtx* zc = ZSTD_createCCtx();
+
+    if (argc != 2) {
+        printf("Usage: seqBench <file>\n"); // TODO provide the block delim option here
+        return 1;
+    }
+
+    FILE *f = fopen(argv[1], "rb");
+    fseek(f, 0, SEEK_END);
+    long inBufSize = ftell(f);
+    fseek(f, 0, SEEK_SET);
+
+    char *inBuf = malloc(inBufSize + 1);
+    fread(inBuf, inBufSize, 1, f);
+    fclose(f);
+
+    // Should work fine for this benchmark, but we really need
+    // a function like ZSTD_compressBound() for sequences
+    size_t seqsSize = 2 * (inBufSize / sizeof(ZSTD_Sequence));
+    ZSTD_Sequence *seqs = (ZSTD_Sequence*)malloc(seqsSize * sizeof(ZSTD_Sequence));
+    char *outBuf = malloc(ZSTD_compressBound(inBufSize));
+
+    ZSTD_generateSequences(zc, seqs, seqsSize, inBuf, inBufSize);
+    ZSTD_CCtx_setParameter(zc, ZSTD_c_blockDelimiters, ZSTD_sf_explicitBlockDelimiters);
+    size_t outBufSize = ZSTD_compressSequences(zc, outBuf, inBufSize, seqs, seqsSize, inBuf, inBufSize);
+    if (ZSTD_isError(outBufSize)) {
+        printf("ERROR: %lu\n", outBufSize);
+        return 1;
+    }
+
+    char *validationBuf = malloc(inBufSize);
+    ZSTD_decompress(validationBuf, inBufSize, outBuf, outBufSize);
+
+    if (memcmp(inBuf, validationBuf, inBufSize) == 0) {
+        printf("Compression and decompression were successful!\n");
+    } else {
+        printf("ERROR: input and validation buffers don't match!\n");
+        for (int i = 0; i < inBufSize; i++) {
+            if (inBuf[i] != validationBuf[i]) {
+                printf("First bad index: %d\n", i);
+                break;
+            }
+        }
+    }
+
+    return 0;
+}