laag-xz 5.2.4.0

data/vendor/git.tukaani.org/xz/src/liblzma/lz/lz_encoder.h

@@ -0,0 +1,327 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file       lz_encoder.h
+/// \brief      LZ in window and match finder API
+///
+//  Authors:    Igor Pavlov
+//              Lasse Collin
+//
+//  This file has been put into the public domain.
+//  You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZ_ENCODER_H
+#define LZMA_LZ_ENCODER_H
+
+#include "common.h"
+
+
+/// A table of these is used by the LZ-based encoder to hold
+/// the length-distance pairs found by the match finder.
+typedef struct {
+	uint32_t len;
+	uint32_t dist;
+} lzma_match;
+
+
+typedef struct lzma_mf_s lzma_mf;
+struct lzma_mf_s {
+	///////////////
+	// In Window //
+	///////////////
+
+	/// Pointer to buffer with data to be compressed
+	uint8_t *buffer;
+
+	/// Total size of the allocated buffer (that is, including all
+	/// the extra space)
+	uint32_t size;
+
+	/// Number of bytes that must be kept available in our input history.
+	/// That is, once keep_size_before bytes have been processed,
+	/// buffer[read_pos - keep_size_before] is the oldest byte that
+	/// must be available for reading.
+	uint32_t keep_size_before;
+
+	/// Number of bytes that must be kept in buffer after read_pos.
+	/// That is, read_pos <= write_pos - keep_size_after as long as
+	/// action is LZMA_RUN; when action != LZMA_RUN, read_pos is allowed
+	/// to reach write_pos so that the last bytes get encoded too.
+	uint32_t keep_size_after;
+
+	/// Match finders store locations of matches using 32-bit integers.
+	/// To avoid adjusting several megabytes of integers every time the
+	/// input window is moved with move_window, we only adjust the
+	/// offset of the buffer. Thus, buffer[value_in_hash_table - offset]
+	/// is the byte pointed by value_in_hash_table.
+	uint32_t offset;
+
+	/// buffer[read_pos] is the next byte to run through the match
+	/// finder. This is incremented in the match finder once the byte
+	/// has been processed.
+	uint32_t read_pos;
+
+	/// Number of bytes that have been ran through the match finder, but
+	/// which haven't been encoded by the LZ-based encoder yet.
+	uint32_t read_ahead;
+
+	/// As long as read_pos is less than read_limit, there is enough
+	/// input available in buffer for at least one encoding loop.
+	///
+	/// Because of the stateful API, read_limit may and will get greater
+	/// than read_pos quite often. This is taken into account when
+	/// calculating the value for keep_size_after.
+	uint32_t read_limit;
+
+	/// buffer[write_pos] is the first byte that doesn't contain valid
+	/// uncompressed data; that is, the next input byte will be copied
+	/// to buffer[write_pos].
+	uint32_t write_pos;
+
+	/// Number of bytes not hashed before read_pos. This is needed to
+	/// restart the match finder after LZMA_SYNC_FLUSH.
+	uint32_t pending;
+
+	//////////////////
+	// Match Finder //
+	//////////////////
+
+	/// Find matches. Returns the number of distance-length pairs written
+	/// to the matches array. This is called only via lzma_mf_find().
+	uint32_t (*find)(lzma_mf *mf, lzma_match *matches);
+
+	/// Skips num bytes. This is like find() but doesn't make the
+	/// distance-length pairs available, thus being a little faster.
+	/// This is called only via mf_skip().
+	void (*skip)(lzma_mf *mf, uint32_t num);
+
+	uint32_t *hash;
+	uint32_t *son;
+	uint32_t cyclic_pos;
+	uint32_t cyclic_size; // Must be dictionary size + 1.
+	uint32_t hash_mask;
+
+	/// Maximum number of loops in the match finder
+	uint32_t depth;
+
+	/// Maximum length of a match that the match finder will try to find.
+	uint32_t nice_len;
+
+	/// Maximum length of a match supported by the LZ-based encoder.
+	/// If the longest match found by the match finder is nice_len,
+	/// mf_find() tries to expand it up to match_len_max bytes.
+	uint32_t match_len_max;
+
+	/// When running out of input, binary tree match finders need to know
+	/// if it is due to flushing or finishing. The action is used also
+	/// by the LZ-based encoders themselves.
+	lzma_action action;
+
+	/// Number of elements in hash[]
+	uint32_t hash_count;
+
+	/// Number of elements in son[]
+	uint32_t sons_count;
+};
+
+
+typedef struct {
+	/// Extra amount of data to keep available before the "actual"
+	/// dictionary.
+	size_t before_size;
+
+	/// Size of the history buffer
+	size_t dict_size;
+
+	/// Extra amount of data to keep available after the "actual"
+	/// dictionary.
+	size_t after_size;
+
+	/// Maximum length of a match that the LZ-based encoder can accept.
+	/// This is used to extend matches of length nice_len to the
+	/// maximum possible length.
+	size_t match_len_max;
+
+	/// Match finder will search matches up to this length.
+	/// This must be less than or equal to match_len_max.
+	size_t nice_len;
+
+	/// Type of the match finder to use
+	lzma_match_finder match_finder;
+
+	/// Maximum search depth
+	uint32_t depth;
+
+	/// TODO: Comment
+	const uint8_t *preset_dict;
+
+	uint32_t preset_dict_size;
+
+} lzma_lz_options;
+
+
+// The total usable buffer space at any moment outside the match finder:
+// before_size + dict_size + after_size + match_len_max
+//
+// In reality, there's some extra space allocated to prevent the number of
+// memmove() calls reasonable. The bigger the dict_size is, the bigger
+// this extra buffer will be since with bigger dictionaries memmove() would
+// also take longer.
+//
+// A single encoder loop in the LZ-based encoder may call the match finder
+// (mf_find() or mf_skip()) at most after_size times. In other words,
+// a single encoder loop may increment lzma_mf.read_pos at most after_size
+// times. Since matches are looked up to
+// lzma_mf.buffer[lzma_mf.read_pos + match_len_max - 1], the total
+// amount of extra buffer needed after dict_size becomes
+// after_size + match_len_max.
+//
+// before_size has two uses. The first one is to keep literals available
+// in cases when the LZ-based encoder has made some read ahead.
+// TODO: Maybe this could be changed by making the LZ-based encoders to
+// store the actual literals as they do with length-distance pairs.
+//
+// Algorithms such as LZMA2 first try to compress a chunk, and then check
+// if the encoded result is smaller than the uncompressed one. If the chunk
+// was uncompressible, it is better to store it in uncompressed form in
+// the output stream. To do this, the whole uncompressed chunk has to be
+// still available in the history buffer. before_size achieves that.
+
+
+typedef struct {
+	/// Data specific to the LZ-based encoder
+	void *coder;
+
+	/// Function to encode from *dict to out[]
+	lzma_ret (*code)(void *coder,
+			lzma_mf *restrict mf, uint8_t *restrict out,
+			size_t *restrict out_pos, size_t out_size);
+
+	/// Free allocated resources
+	void (*end)(void *coder, const lzma_allocator *allocator);
+
+	/// Update the options in the middle of the encoding.
+	lzma_ret (*options_update)(void *coder, const lzma_filter *filter);
+
+} lzma_lz_encoder;
+
+
+// Basic steps:
+//  1. Input gets copied into the dictionary.
+//  2. Data in dictionary gets run through the match finder byte by byte.
+//  3. The literals and matches are encoded using e.g. LZMA.
+//
+// The bytes that have been ran through the match finder, but not encoded yet,
+// are called `read ahead'.
+
+
+/// Get pointer to the first byte not ran through the match finder
+static inline const uint8_t *
+mf_ptr(const lzma_mf *mf)
+{
+	return mf->buffer + mf->read_pos;
+}
+
+
+/// Get the number of bytes that haven't been ran through the match finder yet.
+static inline uint32_t
+mf_avail(const lzma_mf *mf)
+{
+	return mf->write_pos - mf->read_pos;
+}
+
+
+/// Get the number of bytes that haven't been encoded yet (some of these
+/// bytes may have been ran through the match finder though).
+static inline uint32_t
+mf_unencoded(const lzma_mf *mf)
+{
+	return mf->write_pos - mf->read_pos + mf->read_ahead;
+}
+
+
+/// Calculate the absolute offset from the beginning of the most recent
+/// dictionary reset. Only the lowest four bits are important, so there's no
+/// problem that we don't know the 64-bit size of the data encoded so far.
+///
+/// NOTE: When moving the input window, we need to do it so that the lowest
+/// bits of dict->read_pos are not modified to keep this macro working
+/// as intended.
+static inline uint32_t
+mf_position(const lzma_mf *mf)
+{
+	return mf->read_pos - mf->read_ahead;
+}
+
+
+/// Since everything else begins with mf_, use it also for lzma_mf_find().
+#define mf_find lzma_mf_find
+
+
+/// Skip the given number of bytes. This is used when a good match was found.
+/// For example, if mf_find() finds a match of 200 bytes long, the first byte
+/// of that match was already consumed by mf_find(), and the rest 199 bytes
+/// have to be skipped with mf_skip(mf, 199).
+static inline void
+mf_skip(lzma_mf *mf, uint32_t amount)
+{
+	if (amount != 0) {
+		mf->skip(mf, amount);
+		mf->read_ahead += amount;
+	}
+}
+
+
+/// Copies at most *left number of bytes from the history buffer
+/// to out[]. This is needed by LZMA2 to encode uncompressed chunks.
+static inline void
+mf_read(lzma_mf *mf, uint8_t *out, size_t *out_pos, size_t out_size,
+		size_t *left)
+{
+	const size_t out_avail = out_size - *out_pos;
+	const size_t copy_size = my_min(out_avail, *left);
+
+	assert(mf->read_ahead == 0);
+	assert(mf->read_pos >= *left);
+
+	memcpy(out + *out_pos, mf->buffer + mf->read_pos - *left,
+			copy_size);
+
+	*out_pos += copy_size;
+	*left -= copy_size;
+	return;
+}
+
+
+extern lzma_ret lzma_lz_encoder_init(
+		lzma_next_coder *next, const lzma_allocator *allocator,
+		const lzma_filter_info *filters,
+		lzma_ret (*lz_init)(lzma_lz_encoder *lz,
+			const lzma_allocator *allocator, const void *options,
+			lzma_lz_options *lz_options));
+
+
+extern uint64_t lzma_lz_encoder_memusage(const lzma_lz_options *lz_options);
+
+
+// These are only for LZ encoder's internal use.
+extern uint32_t lzma_mf_find(
+		lzma_mf *mf, uint32_t *count, lzma_match *matches);
+
+extern uint32_t lzma_mf_hc3_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_hc3_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_hc4_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_hc4_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_bt2_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_bt2_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_bt3_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_bt3_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_bt4_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_bt4_skip(lzma_mf *dict, uint32_t amount);
+
+#endif

data/vendor/git.tukaani.org/xz/src/liblzma/lz/lz_encoder_hash.h

@@ -0,0 +1,108 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file       lz_encoder_hash.h
+/// \brief      Hash macros for match finders
+//
+//  Author:     Igor Pavlov
+//
+//  This file has been put into the public domain.
+//  You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZ_ENCODER_HASH_H
+#define LZMA_LZ_ENCODER_HASH_H
+
+#if defined(WORDS_BIGENDIAN) && !defined(HAVE_SMALL)
+	// This is to make liblzma produce the same output on big endian
+	// systems that it does on little endian systems. lz_encoder.c
+	// takes care of including the actual table.
+	extern const uint32_t lzma_lz_hash_table[256];
+#	define hash_table lzma_lz_hash_table
+#else
+#	include "check.h"
+#	define hash_table lzma_crc32_table[0]
+#endif
+
+#define HASH_2_SIZE (UINT32_C(1) << 10)
+#define HASH_3_SIZE (UINT32_C(1) << 16)
+#define HASH_4_SIZE (UINT32_C(1) << 20)
+
+#define HASH_2_MASK (HASH_2_SIZE - 1)
+#define HASH_3_MASK (HASH_3_SIZE - 1)
+#define HASH_4_MASK (HASH_4_SIZE - 1)
+
+#define FIX_3_HASH_SIZE (HASH_2_SIZE)
+#define FIX_4_HASH_SIZE (HASH_2_SIZE + HASH_3_SIZE)
+#define FIX_5_HASH_SIZE (HASH_2_SIZE + HASH_3_SIZE + HASH_4_SIZE)
+
+// Endianness doesn't matter in hash_2_calc() (no effect on the output).
+#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
+#	define hash_2_calc() \
+		const uint32_t hash_value = *(const uint16_t *)(cur)
+#else
+#	define hash_2_calc() \
+		const uint32_t hash_value \
+			= (uint32_t)(cur[0]) | ((uint32_t)(cur[1]) << 8)
+#endif
+
+#define hash_3_calc() \
+	const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+	const uint32_t hash_2_value = temp & HASH_2_MASK; \
+	const uint32_t hash_value \
+			= (temp ^ ((uint32_t)(cur[2]) << 8)) & mf->hash_mask
+
+#define hash_4_calc() \
+	const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+	const uint32_t hash_2_value = temp & HASH_2_MASK; \
+	const uint32_t hash_3_value \
+			= (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK; \
+	const uint32_t hash_value = (temp ^ ((uint32_t)(cur[2]) << 8) \
+			^ (hash_table[cur[3]] << 5)) & mf->hash_mask
+
+
+// The following are not currently used.
+
+#define hash_5_calc() \
+	const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+	const uint32_t hash_2_value = temp & HASH_2_MASK; \
+	const uint32_t hash_3_value \
+			= (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK; \
+	uint32_t hash_4_value = (temp ^ ((uint32_t)(cur[2]) << 8) ^ \
+			^ hash_table[cur[3]] << 5); \
+	const uint32_t hash_value \
+			= (hash_4_value ^ (hash_table[cur[4]] << 3)) \
+				& mf->hash_mask; \
+	hash_4_value &= HASH_4_MASK
+
+/*
+#define hash_zip_calc() \
+	const uint32_t hash_value \
+			= (((uint32_t)(cur[0]) | ((uint32_t)(cur[1]) << 8)) \
+				^ hash_table[cur[2]]) & 0xFFFF
+*/
+
+#define hash_zip_calc() \
+	const uint32_t hash_value \
+			= (((uint32_t)(cur[2]) | ((uint32_t)(cur[0]) << 8)) \
+				^ hash_table[cur[1]]) & 0xFFFF
+
+#define mt_hash_2_calc() \
+	const uint32_t hash_2_value \
+			= (hash_table[cur[0]] ^ cur[1]) & HASH_2_MASK
+
+#define mt_hash_3_calc() \
+	const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+	const uint32_t hash_2_value = temp & HASH_2_MASK; \
+	const uint32_t hash_3_value \
+			= (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK
+
+#define mt_hash_4_calc() \
+	const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+	const uint32_t hash_2_value = temp & HASH_2_MASK; \
+	const uint32_t hash_3_value \
+			= (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK; \
+	const uint32_t hash_4_value = (temp ^ ((uint32_t)(cur[2]) << 8) ^ \
+			(hash_table[cur[3]] << 5)) & HASH_4_MASK
+
+#endif

data/vendor/git.tukaani.org/xz/src/liblzma/lz/lz_encoder_hash_table.h

@@ -0,0 +1,68 @@
+/* This file has been automatically generated by crc32_tablegen.c. */
+
+const uint32_t lzma_lz_hash_table[256] = {
+	0x00000000, 0x77073096, 0xEE0E612C, 0x990951BA,
+	0x076DC419, 0x706AF48F, 0xE963A535, 0x9E6495A3,
+	0x0EDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988,
+	0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91,
+	0x1DB71064, 0x6AB020F2, 0xF3B97148, 0x84BE41DE,
+	0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7,
+	0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC,
+	0x14015C4F, 0x63066CD9, 0xFA0F3D63, 0x8D080DF5,
+	0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172,
+	0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B,
+	0x35B5A8FA, 0x42B2986C, 0xDBBBC9D6, 0xACBCF940,
+	0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59,
+	0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116,
+	0x21B4F4B5, 0x56B3C423, 0xCFBA9599, 0xB8BDA50F,
+	0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924,
+	0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D,
+	0x76DC4190, 0x01DB7106, 0x98D220BC, 0xEFD5102A,
+	0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433,
+	0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818,
+	0x7F6A0DBB, 0x086D3D2D, 0x91646C97, 0xE6635C01,
+	0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E,
+	0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457,
+	0x65B0D9C6, 0x12B7E950, 0x8BBEB8EA, 0xFCB9887C,
+	0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65,
+	0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2,
+	0x4ADFA541, 0x3DD895D7, 0xA4D1C46D, 0xD3D6F4FB,
+	0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0,
+	0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7CC9,
+	0x5005713C, 0x270241AA, 0xBE0B1010, 0xC90C2086,
+	0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F,
+	0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4,
+	0x59B33D17, 0x2EB40D81, 0xB7BD5C3B, 0xC0BA6CAD,
+	0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A,
+	0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683,
+	0xE3630B12, 0x94643B84, 0x0D6D6A3E, 0x7A6A5AA8,
+	0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1,
+	0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE,
+	0xF762575D, 0x806567CB, 0x196C3671, 0x6E6B06E7,
+	0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC,
+	0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5,
+	0xD6D6A3E8, 0xA1D1937E, 0x38D8C2C4, 0x4FDFF252,
+	0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B,
+	0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60,
+	0xDF60EFC3, 0xA867DF55, 0x316E8EEF, 0x4669BE79,
+	0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236,
+	0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F,
+	0xC5BA3BBE, 0xB2BD0B28, 0x2BB45A92, 0x5CB36A04,
+	0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D,
+	0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A,
+	0x9C0906A9, 0xEB0E363F, 0x72076785, 0x05005713,
+	0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38,
+	0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21,
+	0x86D3D2D4, 0xF1D4E242, 0x68DDB3F8, 0x1FDA836E,
+	0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777,
+	0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C,
+	0x8F659EFF, 0xF862AE69, 0x616BFFD3, 0x166CCF45,
+	0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2,
+	0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB,
+	0xAED16A4A, 0xD9D65ADC, 0x40DF0B66, 0x37D83BF0,
+	0xA9BCAE53, 0xDEBB9EC5, 0x47B2CF7F, 0x30B5FFE9,
+	0xBDBDF21C, 0xCABAC28A, 0x53B39330, 0x24B4A3A6,
+	0xBAD03605, 0xCDD70693, 0x54DE5729, 0x23D967BF,
+	0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94,
+	0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D
+};