zstd-native-ruby 1.0.0

data/vendor/zstd/contrib/linux-kernel/zstd_decompress_module.c

@@ -0,0 +1,141 @@
+// SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * 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).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/zstd.h>
+
+#include "common/zstd_deps.h"
+
+/* Common symbols. zstd_compress must depend on zstd_decompress. */
+
+unsigned int zstd_is_error(size_t code)
+{
+	return ZSTD_isError(code);
+}
+EXPORT_SYMBOL(zstd_is_error);
+
+zstd_error_code zstd_get_error_code(size_t code)
+{
+	return ZSTD_getErrorCode(code);
+}
+EXPORT_SYMBOL(zstd_get_error_code);
+
+const char *zstd_get_error_name(size_t code)
+{
+	return ZSTD_getErrorName(code);
+}
+EXPORT_SYMBOL(zstd_get_error_name);
+
+/* Decompression symbols. */
+
+size_t zstd_dctx_workspace_bound(void)
+{
+	return ZSTD_estimateDCtxSize();
+}
+EXPORT_SYMBOL(zstd_dctx_workspace_bound);
+
+zstd_dctx *zstd_create_dctx_advanced(zstd_custom_mem custom_mem)
+{
+	return ZSTD_createDCtx_advanced(custom_mem);
+}
+EXPORT_SYMBOL(zstd_create_dctx_advanced);
+
+size_t zstd_free_dctx(zstd_dctx *dctx)
+{
+	return ZSTD_freeDCtx(dctx);
+}
+EXPORT_SYMBOL(zstd_free_dctx);
+
+zstd_ddict *zstd_create_ddict_byreference(const void *dict, size_t dict_size,
+					  zstd_custom_mem custom_mem)
+{
+	return ZSTD_createDDict_advanced(dict, dict_size, ZSTD_dlm_byRef,
+					 ZSTD_dct_auto, custom_mem);
+
+}
+EXPORT_SYMBOL(zstd_create_ddict_byreference);
+
+size_t zstd_free_ddict(zstd_ddict *ddict)
+{
+	return ZSTD_freeDDict(ddict);
+}
+EXPORT_SYMBOL(zstd_free_ddict);
+
+zstd_dctx *zstd_init_dctx(void *workspace, size_t workspace_size)
+{
+	if (workspace == NULL)
+		return NULL;
+	return ZSTD_initStaticDCtx(workspace, workspace_size);
+}
+EXPORT_SYMBOL(zstd_init_dctx);
+
+size_t zstd_decompress_dctx(zstd_dctx *dctx, void *dst, size_t dst_capacity,
+	const void *src, size_t src_size)
+{
+	return ZSTD_decompressDCtx(dctx, dst, dst_capacity, src, src_size);
+}
+EXPORT_SYMBOL(zstd_decompress_dctx);
+
+size_t zstd_decompress_using_ddict(zstd_dctx *dctx,
+	void *dst, size_t dst_capacity, const void* src, size_t src_size,
+	const zstd_ddict* ddict)
+{
+	return ZSTD_decompress_usingDDict(dctx, dst, dst_capacity, src,
+					  src_size, ddict);
+}
+EXPORT_SYMBOL(zstd_decompress_using_ddict);
+
+size_t zstd_dstream_workspace_bound(size_t max_window_size)
+{
+	return ZSTD_estimateDStreamSize(max_window_size);
+}
+EXPORT_SYMBOL(zstd_dstream_workspace_bound);
+
+zstd_dstream *zstd_init_dstream(size_t max_window_size, void *workspace,
+	size_t workspace_size)
+{
+	if (workspace == NULL)
+		return NULL;
+	(void)max_window_size;
+	return ZSTD_initStaticDStream(workspace, workspace_size);
+}
+EXPORT_SYMBOL(zstd_init_dstream);
+
+size_t zstd_reset_dstream(zstd_dstream *dstream)
+{
+	return ZSTD_DCtx_reset(dstream, ZSTD_reset_session_only);
+}
+EXPORT_SYMBOL(zstd_reset_dstream);
+
+size_t zstd_decompress_stream(zstd_dstream *dstream, zstd_out_buffer *output,
+	zstd_in_buffer *input)
+{
+	return ZSTD_decompressStream(dstream, output, input);
+}
+EXPORT_SYMBOL(zstd_decompress_stream);
+
+size_t zstd_find_frame_compressed_size(const void *src, size_t src_size)
+{
+	return ZSTD_findFrameCompressedSize(src, src_size);
+}
+EXPORT_SYMBOL(zstd_find_frame_compressed_size);
+
+size_t zstd_get_frame_header(zstd_frame_header *header, const void *src,
+	size_t src_size)
+{
+	return ZSTD_getFrameHeader(header, src, src_size);
+}
+EXPORT_SYMBOL(zstd_get_frame_header);
+
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_DESCRIPTION("Zstd Decompressor");

data/vendor/zstd/contrib/linux-kernel/zstd_deps.h

@@ -0,0 +1,121 @@
+/* SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause */
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * 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).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+/*
+ * This file provides common libc dependencies that zstd requires.
+ * The purpose is to allow replacing this file with a custom implementation
+ * to compile zstd without libc support.
+ */
+
+/* Need:
+ * NULL
+ * INT_MAX
+ * UINT_MAX
+ * ZSTD_memcpy()
+ * ZSTD_memset()
+ * ZSTD_memmove()
+ */
+#ifndef ZSTD_DEPS_COMMON
+#define ZSTD_DEPS_COMMON
+
+#include <linux/limits.h>
+#include <linux/stddef.h>
+
+#define ZSTD_memcpy(d,s,n) __builtin_memcpy((d),(s),(n))
+#define ZSTD_memmove(d,s,n) __builtin_memmove((d),(s),(n))
+#define ZSTD_memset(d,s,n) __builtin_memset((d),(s),(n))
+
+#endif /* ZSTD_DEPS_COMMON */
+
+/*
+ * Define malloc as always failing. That means the user must
+ * either use ZSTD_customMem or statically allocate memory.
+ * Need:
+ * ZSTD_malloc()
+ * ZSTD_free()
+ * ZSTD_calloc()
+ */
+#ifdef ZSTD_DEPS_NEED_MALLOC
+#ifndef ZSTD_DEPS_MALLOC
+#define ZSTD_DEPS_MALLOC
+
+#define ZSTD_malloc(s) ({ (void)(s); NULL; })
+#define ZSTD_free(p) ((void)(p))
+#define ZSTD_calloc(n,s) ({ (void)(n); (void)(s); NULL; })
+
+#endif /* ZSTD_DEPS_MALLOC */
+#endif /* ZSTD_DEPS_NEED_MALLOC */
+
+/*
+ * Provides 64-bit math support.
+ * Need:
+ * U64 ZSTD_div64(U64 dividend, U32 divisor)
+ */
+#ifdef ZSTD_DEPS_NEED_MATH64
+#ifndef ZSTD_DEPS_MATH64
+#define ZSTD_DEPS_MATH64
+
+#include <linux/math64.h>
+
+static uint64_t ZSTD_div64(uint64_t dividend, uint32_t divisor) {
+  return div_u64(dividend, divisor);
+}
+
+#endif /* ZSTD_DEPS_MATH64 */
+#endif /* ZSTD_DEPS_NEED_MATH64 */
+
+/*
+ * This is only requested when DEBUGLEVEL >= 1, meaning
+ * it is disabled in production.
+ * Need:
+ * assert()
+ */
+#ifdef ZSTD_DEPS_NEED_ASSERT
+#ifndef ZSTD_DEPS_ASSERT
+#define ZSTD_DEPS_ASSERT
+
+#include <linux/kernel.h>
+
+#define assert(x) WARN_ON(!(x))
+
+#endif /* ZSTD_DEPS_ASSERT */
+#endif /* ZSTD_DEPS_NEED_ASSERT */
+
+/*
+ * This is only requested when DEBUGLEVEL >= 2, meaning
+ * it is disabled in production.
+ * Need:
+ * ZSTD_DEBUG_PRINT()
+ */
+#ifdef ZSTD_DEPS_NEED_IO
+#ifndef ZSTD_DEPS_IO
+#define ZSTD_DEPS_IO
+
+#include <linux/printk.h>
+
+#define ZSTD_DEBUG_PRINT(...) pr_debug(__VA_ARGS__)
+
+#endif /* ZSTD_DEPS_IO */
+#endif /* ZSTD_DEPS_NEED_IO */
+
+/*
+ * Only requested when MSAN is enabled.
+ * Need:
+ * intptr_t
+ */
+#ifdef ZSTD_DEPS_NEED_STDINT
+#ifndef ZSTD_DEPS_STDINT
+#define ZSTD_DEPS_STDINT
+
+/* intptr_t already provided by ZSTD_DEPS_COMMON */
+
+#endif /* ZSTD_DEPS_STDINT */
+#endif /* ZSTD_DEPS_NEED_STDINT */

data/vendor/zstd/contrib/match_finders/README.md

@@ -0,0 +1,42 @@
+## Edit Distance Match Finder
+
+```
+/* This match finder leverages techniques used in file comparison algorithms
+ * to find matches between a dictionary and a source file.
+ * 
+ * The original motivation for studying this approach was to try and optimize 
+ * Zstandard for the use case of patching: the most common scenario being 
+ * updating an existing software package with the next version. When patching,
+ * the difference between the old version of the package and the new version 
+ * is generally tiny (most of the new file will be identical to 
+ * the old one). In more technical terms, the edit distance (the minimal number 
+ * of changes required to take one sequence of bytes to another) between the 
+ * files would be small relative to the size of the file. 
+ * 
+ * Various 'diffing' algorithms utilize this notion of edit distance and 
+ * the corresponding concept of a minimal edit script between two 
+ * sequences to identify the regions within two files where they differ. 
+ * The core algorithm used in this match finder is described in: 
+ * 
+ * "An O(ND) Difference Algorithm and its Variations", Eugene W. Myers,
+ *    Algorithmica Vol. 1, 1986, pp. 251-266,
+ *    <https://doi.org/10.1007/BF01840446>.
+ * 
+ * Additional algorithmic heuristics for speed improvement have also been included.
+ * These we inspired from implementations of various regular and binary diffing 
+ * algorithms such as GNU diff, bsdiff, and Xdelta. 
+ * 
+ * Note: after some experimentation, this approach proved to not provide enough 
+ * utility to justify the additional CPU used in finding matches. The one area
+ * where this approach consistently outperforms Zstandard even on level 19 is 
+ * when compressing small files (<10 KB) using an equally small dictionary that 
+ * is very similar to the source file. For the use case that this was intended,
+ * (large similar files) this approach by itself took 5-10X longer than zstd-19 and 
+ * generally resulted in 2-3X larger files. The core advantage that zstd-19 has 
+ * over this approach for match finding is the overlapping matches. This approach 
+ * cannot find any. 
+ * 
+ * I'm leaving this in the contrib section in case this ever becomes interesting 
+ * to explore again.
+ * */
+```