X-Git-Url: https://wimlib.net/git/?p=wimlib;a=blobdiff_plain;f=src%2Flzx-compress.c;h=b337697eda3abb8688d33ddc4c14863f37dbfd25;hp=a31fc2ab2271e889f5e9cd2bde4f2657bf5b8a0d;hb=5c1b50fdca21aaaddbdb34f5f7beec7d67016984;hpb=9bec19341aac9190cd0bd80a6937b02d923dd566 diff --git a/src/lzx-compress.c b/src/lzx-compress.c index a31fc2ab..b337697e 100644 --- a/src/lzx-compress.c +++ b/src/lzx-compress.c @@ -42,7 +42,8 @@ * and certain other details are quite similar, such as the method for storing * Huffman codes. However, some of the main differences are: * - * - LZX preprocesses the data before attempting to compress it. + * - LZX preprocesses the data to attempt to make x86 machine code slightly more + * compressible before attempting to compress it further. * - LZX uses a "main" alphabet which combines literals and matches, with the * match symbols containing a "length header" (giving all or part of the match * length) and a "position slot" (giving, roughly speaking, the order of @@ -51,7 +52,8 @@ * dynamic Huffman blocks ("aligned offset" and "verbatim"). * - LZX has a minimum match length of 2 rather than 3. * - In LZX, match offsets 0 through 2 actually represent entries in an LRU - * queue of match offsets. + * queue of match offsets. This is very useful for certain types of files, + * such as binary files that have repeating records. * * Algorithms * ========== @@ -66,34 +68,39 @@ * * The "slow" algorithm to generate LZX-compressed data is roughly as follows: * - * 1. Preprocess the input data to translate the targets of x86 call instructions - * to absolute offsets. + * 1. Preprocess the input data to translate the targets of x86 call + * instructions to absolute offsets. * - * 2. Build the suffix array and inverse suffix array for the input data. + * 2. Build the suffix array and inverse suffix array for the input data. The + * suffix array contains the indices of all suffixes of the input data, + * sorted lexcographically by the corresponding suffixes. The "position" of + * a suffix is the index of that suffix in the original string, whereas the + * "rank" of a suffix is the index at which that suffix's position is found + * in the suffix array. * * 3. Build the longest common prefix array corresponding to the suffix array. * - * 4. For each suffix rank, find the highest lower suffix rank that has a - * lower position, the lowest higher suffix rank that has a lower position, - * and the length of the common prefix shared between each. (Position = - * index of suffix in original string, rank = index of suffix in suffix - * array.) This information is later used to link suffix ranks into a - * doubly-linked list for searching the suffix array. + * 4. For each suffix, find the highest lower ranked suffix that has a lower + * position, the lowest higher ranked suffix that has a lower position, and + * the length of the common prefix shared between each. This information is + * later used to link suffix ranks into a doubly-linked list for searching + * the suffix array. * * 5. Set a default cost model for matches/literals. * - * 6. Determine the lowest cost sequence of LZ77 matches ((offset, length) pairs) - * and literal bytes to divide the input into. Raw match-finding is done by - * searching the suffix array using a linked list to avoid considering any - * suffixes that start after the current position. Each run of the - * match-finder returns the lowest-cost longest match as well as any shorter - * matches that have even lower costs. Each such run also adds the suffix - * rank of the current position into the linked list being used to search the - * suffix array. Parsing, or match-choosing, is solved as a minimum-cost - * path problem using a forward "optimal parsing" algorithm based on the - * Deflate encoder from 7-Zip. This algorithm moves forward calculating the - * minimum cost to reach each byte until either a very long match is found or - * until a position is found at which no matches start or overlap. + * 6. Determine the lowest cost sequence of LZ77 matches ((offset, length) + * pairs) and literal bytes to divide the input into. Raw match-finding is + * done by searching the suffix array using a linked list to avoid + * considering any suffixes that start after the current position. Each run + * of the match-finder returns the approximate lowest-cost longest match as + * well as any shorter matches that have even lower approximate costs. Each + * such run also adds the suffix rank of the current position into the linked + * list being used to search the suffix array. Parsing, or match-choosing, + * is solved as a minimum-cost path problem using a forward "optimal parsing" + * algorithm based on the Deflate encoder from 7-Zip. This algorithm moves + * forward calculating the minimum cost to reach each byte until either a + * very long match is found or until a position is found at which no matches + * start or overlap. * * 7. Build the Huffman codes needed to output the matches/literals. * @@ -104,45 +111,18 @@ * 9. Output the resulting block using the match/literal sequences and the * Huffman codes that were computed for the block. * + * Note: the algorithm does not yet attempt to split the input into multiple LZX + * blocks; it instead uses a series of blocks of LZX_DIV_BLOCK_SIZE bytes. + * * Fast algorithm * -------------- * * The fast algorithm (and the only one available in wimlib v1.5.1 and earlier) * spends much less time on the main bottlenecks of the compression process --- - * that is the match finding, match choosing, and block splitting. Matches are - * found and chosen with hash chains using a greedy parse with one position of - * look-ahead. No block splitting is done; only compressing the full input into - * an aligned offset block is considered. - * - * API - * === - * - * The old API (retained for backward compatibility) consists of just one function: - * - * wimlib_lzx_compress() - * - * The new compressor has more potential parameters and needs more memory, so - * the new API ties up memory allocations and compression parameters into a - * context: - * - * wimlib_lzx_alloc_context() - * wimlib_lzx_compress2() - * wimlib_lzx_free_context() - * - * Both wimlib_lzx_compress() and wimlib_lzx_compress2() are designed to - * compress an in-memory buffer of up to 32768 bytes. There is no sliding - * window. This is suitable for the WIM format, which uses fixed-size chunks - * that are seemingly always 32768 bytes. If needed, the compressor potentially - * could be extended to support a larger and/or sliding window. - * - * Both wimlib_lzx_compress() and wimlib_lzx_compress2() return 0 if the data - * could not be compressed to less than the size of the uncompressed data. - * Again, this is suitable for the WIM format, which stores such data chunks - * uncompressed. - * - * The functions in this API are exported from the library, although this is - * only in case other programs happen to have uses for it other than WIM - * reading/writing as already handled through the rest of the library. + * that is, the match finding and match choosing. Matches are found and chosen + * with hash chains using a greedy parse with one position of look-ahead. No + * block splitting is done; only compressing the full input into an aligned + * offset block is considered. * * Acknowledgments * =============== @@ -150,7 +130,8 @@ * Acknowledgments to several open-source projects and research papers that made * it possible to implement this code: * - * - divsufsort (author: Yuta Mori), for the suffix array construction code. + * - divsufsort (author: Yuta Mori), for the suffix array construction code, + * located in a separate directory (divsufsort/). * * - "Linear-Time Longest-Common-Prefix Computation in Suffix Arrays and Its * Applications" (Kasai et al. 2001), for the LCP array computation. @@ -162,7 +143,7 @@ * (match-choosing). * * - zlib (author: Jean-loup Gailly and Mark Adler), for the hash table - * match-finding algorithm. + * match-finding algorithm (used in lz77.c). * * - lzx-compress (author: Matthew T. Russotto), on which some parts of this * code were originally based. @@ -173,41 +154,32 @@ #endif #include "wimlib.h" -#include "wimlib/compress.h" +#include "wimlib/compressor_ops.h" +#include "wimlib/compress_common.h" +#include "wimlib/endianness.h" #include "wimlib/error.h" +#include "wimlib/lz_hash.h" +#include "wimlib/lz_sarray.h" #include "wimlib/lzx.h" #include "wimlib/util.h" -#include -#include #include #ifdef ENABLE_LZX_DEBUG -# include +# include "wimlib/decompress_common.h" #endif -#include "divsufsort/divsufsort.h" - -typedef freq_t input_idx_t; -typedef u32 sym_cost_t; typedef u32 block_cost_t; -#define INFINITE_SYM_COST ((sym_cost_t)~0U) -#define INFINITE_BLOCK_COST ((block_cost_t)~0U) +#define INFINITE_BLOCK_COST (~(block_cost_t)0) #define LZX_OPTIM_ARRAY_SIZE 4096 -/* Currently, this constant can't simply be changed because the code currently - * uses a static number of position slots (and may make other assumptions as - * well). */ -#define LZX_MAX_WINDOW_SIZE 32768 - -/* This may be WIM-specific */ -#define LZX_DEFAULT_BLOCK_SIZE 32768 +#define LZX_DIV_BLOCK_SIZE 32768 #define LZX_MAX_CACHE_PER_POS 10 /* Codewords for the LZX main, length, and aligned offset Huffman codes */ struct lzx_codewords { - u16 main[LZX_MAINCODE_NUM_SYMBOLS]; + u16 main[LZX_MAINCODE_MAX_NUM_SYMBOLS]; u16 len[LZX_LENCODE_NUM_SYMBOLS]; u16 aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS]; }; @@ -218,7 +190,7 @@ struct lzx_codewords { * A 0 length means the codeword has zero frequency. */ struct lzx_lens { - u8 main[LZX_MAINCODE_NUM_SYMBOLS]; + u8 main[LZX_MAINCODE_MAX_NUM_SYMBOLS]; u8 len[LZX_LENCODE_NUM_SYMBOLS]; u8 aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS]; }; @@ -229,9 +201,9 @@ struct lzx_lens { * --- generally a high cost, since even if it gets used in the next iteration, * it probably will not be used very times. */ struct lzx_costs { - sym_cost_t main[LZX_MAINCODE_NUM_SYMBOLS]; - sym_cost_t len[LZX_LENCODE_NUM_SYMBOLS]; - sym_cost_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS]; + u8 main[LZX_MAINCODE_MAX_NUM_SYMBOLS]; + u8 len[LZX_LENCODE_NUM_SYMBOLS]; + u8 aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS]; }; /* The LZX main, length, and aligned offset Huffman codes */ @@ -242,9 +214,9 @@ struct lzx_codes { /* Tables for tallying symbol frequencies in the three LZX alphabets */ struct lzx_freqs { - freq_t main[LZX_MAINCODE_NUM_SYMBOLS]; - freq_t len[LZX_LENCODE_NUM_SYMBOLS]; - freq_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS]; + input_idx_t main[LZX_MAINCODE_MAX_NUM_SYMBOLS]; + input_idx_t len[LZX_LENCODE_NUM_SYMBOLS]; + input_idx_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS]; }; /* LZX intermediate match/literal format */ @@ -258,25 +230,13 @@ struct lzx_match { * * 8-24 position footer. This is the offset of the real formatted * offset from the position base. This can be at most 17 bits - * (since lzx_extra_bits[LZX_NUM_POSITION_SLOTS - 1] is 17). + * (since lzx_extra_bits[LZX_MAX_POSITION_SLOTS - 1] is 17). * * 0-7 length of match, minus 2. This can be at most * (LZX_MAX_MATCH_LEN - 2) == 255, so it will fit in 8 bits. */ u32 data; }; -/* Raw LZ match/literal format: just a length and offset. - * - * The length is the number of bytes of the match, and the offset is the number - * of bytes back in the input the match is from the current position. - * - * If @len < LZX_MIN_MATCH_LEN, then it's really just a literal byte and @offset is - * meaningless. */ -struct raw_match { - u16 len; - input_idx_t offset; -}; - /* Specification for an LZX block. */ struct lzx_block_spec { @@ -302,103 +262,44 @@ struct lzx_block_spec { struct lzx_codes codes; }; -/* - * An array of these structures is used during the match-choosing algorithm. - * They correspond to consecutive positions in the window and are used to keep - * track of the cost to reach each position, and the match/literal choices that - * need to be chosen to reach that position. - */ -struct lzx_optimal { - /* The approximate minimum cost, in bits, to reach this position in the - * window which has been found so far. */ - block_cost_t cost; - - /* The union here is just for clarity, since the fields are used in two - * slightly different ways. Initially, the @prev structure is filled in - * first, and links go from later in the window to earlier in the - * window. Later, @next structure is filled in and links go from - * earlier in the window to later in the window. */ - union { - struct { - /* Position of the start of the match or literal that - * was taken to get to this position in the approximate - * minimum-cost parse. */ - input_idx_t link; - - /* Offset (as in an LZ (length, offset) pair) of the - * match or literal that was taken to get to this - * position in the approximate minimum-cost parse. */ - input_idx_t match_offset; - } prev; - struct { - /* Position at which the match or literal starting at - * this position ends in the minimum-cost parse. */ - input_idx_t link; - - /* Offset (as in an LZ (length, offset) pair) of the - * match or literal starting at this position in the - * approximate minimum-cost parse. */ - input_idx_t match_offset; - } next; - }; - - /* The match offset LRU queue that will exist when the approximate - * minimum-cost path to reach this position is taken. */ - struct lzx_lru_queue queue; -}; - -/* Suffix array link */ -struct salink { - /* Rank of highest ranked suffix that has rank lower than the suffix - * corresponding to this structure and either has a lower position - * (initially) or has a position lower than the highest position at - * which matches have been searched for so far, or -1 if there is no - * such suffix. */ - input_idx_t prev; - - /* Rank of lowest ranked suffix that has rank greater than the suffix - * corresponding to this structure and either has a lower position - * (intially) or has a position lower than the highest position at which - * matches have been searched for so far, or -1 if there is no such - * suffix. */ - input_idx_t next; - - /* Length of longest common prefix between the suffix corresponding to - * this structure and the suffix with rank @prev, or 0 if @prev is -1. - */ - input_idx_t lcpprev; - - /* Length of longest common prefix between the suffix corresponding to - * this structure and the suffix with rank @next, or 0 if @next is -1. - */ - input_idx_t lcpnext; -}; +/* Include template for the match-choosing algorithm. */ +#define LZ_COMPRESSOR struct lzx_compressor +#define LZ_ADAPTIVE_STATE struct lzx_lru_queue +struct lzx_compressor; +#include "wimlib/lz_optimal.h" /* State of the LZX compressor. */ struct lzx_compressor { /* The parameters that were used to create the compressor. */ - struct wimlib_lzx_params params; + struct wimlib_lzx_compressor_params params; /* The buffer of data to be compressed. * * 0xe8 byte preprocessing is done directly on the data here before * further compression. * - * Note that this compressor does *not* use a sliding window!!!! It's - * not needed in the WIM format, since every chunk is compressed + * Note that this compressor does *not* use a real sliding window!!!! + * It's not needed in the WIM format, since every chunk is compressed * independently. This is by design, to allow random access to the * chunks. * * We reserve a few extra bytes to potentially allow reading off the end - * of the array in the match-finding code for optimization purposes. - */ - u8 window[LZX_MAX_WINDOW_SIZE + 12]; + * of the array in the match-finding code for optimization purposes + * (currently only needed for the hash chain match-finder). */ + u8 *window; /* Number of bytes of data to be compressed, which is the number of * bytes of data in @window that are actually valid. */ input_idx_t window_size; + /* Allocated size of the @window. */ + input_idx_t max_window_size; + + /* Number of symbols in the main alphabet (depends on the + * @max_window_size since it determines the maximum allowed offset). */ + unsigned num_main_syms; + /* The current match offset LRU queue. */ struct lzx_lru_queue queue; @@ -406,10 +307,6 @@ struct lzx_compressor { * block. */ struct lzx_match *chosen_matches; - struct raw_match *cached_matches; - unsigned cached_matches_pos; - bool matches_cached; - /* Information about the LZX blocks the preprocessed input was divided * into. */ struct lzx_block_spec *block_specs; @@ -424,110 +321,39 @@ struct lzx_compressor { * codewords. */ struct lzx_codes zero_codes; - /* Slow algorithm only: The current cost model. */ + /* The current cost model. */ struct lzx_costs costs; - /* Slow algorithm only: Suffix array for window. - * This is a mapping from suffix rank to suffix position. - * - * Suffix rank means the index of the suffix in the sorted list of - * suffixes, whereas suffix position means the index in the string at - * which the suffix starts. - */ - input_idx_t *SA; - - /* Slow algorithm only: Inverse suffix array for window. - * This is a mapping from suffix position to suffix rank. - * In other words, if 0 <= r < window_size, then ISA[SA[r]] == r. */ - input_idx_t *ISA; + /* Fast algorithm only: Array of hash table links. */ + input_idx_t *prev_tab; - /* Slow algorithm only: Longest Common Prefix array. LCP[i] is the - * number of initial bytes that the suffixes at positions SA[i - 1] and - * SA[i] share. LCP[0] is undefined. */ - input_idx_t *LCP; + /* Slow algorithm only: Suffix array match-finder. */ + struct lz_sarray lz_sarray; - /* Slow algorithm only: Suffix array links. - * - * During a linear scan of the input string to find matches, this array - * used to keep track of which rank suffixes in the suffix array appear - * before the current position. Instead of searching in the original - * suffix array, scans for matches at a given position traverse a linked - * list containing only suffixes that appear before that position. */ - struct salink *salink; - - /* Slow algorithm only: Position in window of next match to return. - * This cannot simply be modified, as the match-finder must still be - * synchronized on the same position. To seek forwards or backwards, - * use lzx_lz_skip_bytes() or lzx_lz_rewind_matchfinder(), respectively. - */ + /* Position in window of next match to return. */ input_idx_t match_window_pos; - /* Slow algorithm only: The match-finder shall ensure the length of - * matches does not exceed this position in the input. */ + /* The match-finder shall ensure the length of matches does not exceed + * this position in the input. */ input_idx_t match_window_end; - /* Slow algorithm only: Temporary space used for match-choosing - * algorithm. - * - * The size of this array must be at least LZX_MAX_MATCH_LEN but - * otherwise is arbitrary. More space simply allows the match-choosing - * algorithm to potentially find better matches (depending on the input, - * as always). */ - struct lzx_optimal *optimum; + /* Matches found by the match-finder are cached in the following array + * to achieve a slight speedup when the same matches are needed on + * subsequent passes. This is suboptimal because different matches may + * be preferred with different cost models, but seems to be a worthwhile + * speedup. */ + struct raw_match *cached_matches; + unsigned cached_matches_pos; + bool matches_cached; - /* Slow algorithm only: Variables used by the match-choosing algorithm. - * - * When matches have been chosen, optimum_cur_idx is set to the position - * in the window of the next match/literal to return and optimum_end_idx - * is set to the position in the window at the end of the last - * match/literal to return. */ - u32 optimum_cur_idx; - u32 optimum_end_idx; + /* Match chooser. */ + struct lz_match_chooser mc; }; -/* Returns the LZX position slot that corresponds to a given formatted offset. - * - * Logically, this returns the smallest i such that - * formatted_offset >= lzx_position_base[i]. - * - * The actual implementation below takes advantage of the regularity of the - * numbers in the lzx_position_base array to calculate the slot directly from - * the formatted offset without actually looking at the array. - */ -static _always_inline_attribute unsigned -lzx_get_position_slot_raw(unsigned formatted_offset) -{ -#if 0 - /* - * Slots 36-49 (formatted_offset >= 262144) can be found by - * (formatted_offset/131072) + 34 == (formatted_offset >> 17) + 34; - * however, this check for formatted_offset >= 262144 is commented out - * because WIM chunks cannot be that large. - */ - if (formatted_offset >= 262144) { - return (formatted_offset >> 17) + 34; - } else -#endif - { - /* Note: this part here only works if: - * - * 2 <= formatted_offset < 655360 - * - * It is < 655360 because the frequency of the position bases - * increases starting at the 655360 entry, and it is >= 2 - * because the below calculation fails if the most significant - * bit is lower than the 2's place. */ - LZX_ASSERT(2 <= formatted_offset && formatted_offset < 655360); - unsigned mssb_idx = bsr32(formatted_offset); - return (mssb_idx << 1) | - ((formatted_offset >> (mssb_idx - 1)) & 1); - } -} - - /* Returns the LZX position slot that corresponds to a given match offset, - * taking into account the recent offset queue (and optionally updating it). */ -static _always_inline_attribute unsigned + * taking into account the recent offset queue and updating it if the offset is + * found in it. */ +static unsigned lzx_get_position_slot(unsigned offset, struct lzx_lru_queue *queue) { unsigned position_slot; @@ -542,9 +368,9 @@ lzx_get_position_slot(unsigned offset, struct lzx_lru_queue *queue) * LRU queue because repeat matches are simply * swapped to the front. */ swap(queue->R[0], queue->R[i]); - /* For recent offsets, the position slot is simply the - * index of the entry in the queue. */ + /* The resulting position slot is simply the first index + * at which the offset was found in the queue. */ return i; } } @@ -566,9 +392,10 @@ lzx_get_position_slot(unsigned offset, struct lzx_lru_queue *queue) * a set of tables that map symbols to codewords and codeword lengths. */ static void lzx_make_huffman_codes(const struct lzx_freqs *freqs, - struct lzx_codes *codes) + struct lzx_codes *codes, + unsigned num_main_syms) { - make_canonical_huffman_code(LZX_MAINCODE_NUM_SYMBOLS, + make_canonical_huffman_code(num_main_syms, LZX_MAX_MAIN_CODEWORD_LEN, freqs->main, codes->lens.main, @@ -588,13 +415,18 @@ lzx_make_huffman_codes(const struct lzx_freqs *freqs, } /* - * Output an LZX match. - * - * @out: The bitstream to write the match to. - * @block_type: The type of the LZX block (LZX_BLOCKTYPE_ALIGNED or LZX_BLOCKTYPE_VERBATIM) - * @match: The match. - * @codes: Pointer to a structure that contains the codewords for the - * main, length, and aligned offset Huffman codes. + * Output a precomputed LZX match. + * + * @out: + * The bitstream to which to write the match. + * @block_type: + * The type of the LZX block (LZX_BLOCKTYPE_ALIGNED or + * LZX_BLOCKTYPE_VERBATIM) + * @match: + * The match, as a (length, offset) pair. + * @codes: + * Pointer to a structure that contains the codewords for the main, length, + * and aligned offset Huffman codes for the current LZX compressed block. */ static void lzx_write_match(struct output_bitstream *out, int block_type, @@ -633,7 +465,7 @@ lzx_write_match(struct output_bitstream *out, int block_type, } /* Combine the position slot with the length header into a single symbol - * that will be encoded with the main tree. + * that will be encoded with the main code. * * The actual main symbol is offset by LZX_NUM_CHARS because values * under LZX_NUM_CHARS are used to indicate a literal byte rather than a @@ -655,7 +487,7 @@ lzx_write_match(struct output_bitstream *out, int block_type, /* For aligned offset blocks with at least 3 extra bits, output the * verbatim bits literally, then the aligned bits encoded using the - * aligned offset tree. Otherwise, only the verbatim bits need to be + * aligned offset code. Otherwise, only the verbatim bits need to be * output. */ if ((block_type == LZX_BLOCKTYPE_ALIGNED) && (num_extra_bits >= 3)) { @@ -674,11 +506,21 @@ lzx_write_match(struct output_bitstream *out, int block_type, } } +/* Output an LZX literal (encoded with the main Huffman code). */ +static void +lzx_write_literal(struct output_bitstream *out, u8 literal, + const struct lzx_codes *codes) +{ + bitstream_put_bits(out, + codes->codewords.main[literal], + codes->lens.main[literal]); +} + static unsigned lzx_build_precode(const u8 lens[restrict], const u8 prev_lens[restrict], const unsigned num_syms, - freq_t precode_freqs[restrict LZX_PRECODE_NUM_SYMBOLS], + input_idx_t precode_freqs[restrict LZX_PRECODE_NUM_SYMBOLS], u8 output_syms[restrict num_syms], u8 precode_lens[restrict LZX_PRECODE_NUM_SYMBOLS], u16 precode_codewords[restrict LZX_PRECODE_NUM_SYMBOLS], @@ -694,7 +536,7 @@ lzx_build_precode(const u8 lens[restrict], * literally. * * output_syms[] will be filled in with the length symbols that will be - * output, including RLE codes, not yet encoded using the pre-tree. + * output, including RLE codes, not yet encoded using the precode. * * cur_run_len keeps track of how many code word lengths are in the * current run of identical lengths. */ @@ -760,7 +602,7 @@ lzx_build_precode(const u8 lens[restrict], * * The extra length symbol is encoded as a difference * from the length of the codeword for the first symbol - * in the run in the previous tree. + * in the run in the previous code. * */ while (cur_run_len >= 4) { unsigned additional_bits; @@ -783,7 +625,7 @@ lzx_build_precode(const u8 lens[restrict], /* Any remaining lengths in the run are outputted without RLE, * as a difference from the length of that codeword in the - * previous tree. */ + * previous code. */ while (cur_run_len > 0) { signed char delta; @@ -813,23 +655,33 @@ lzx_build_precode(const u8 lens[restrict], } /* - * Writes a compressed Huffman code to the output, preceded by the precode for - * it. - * - * The Huffman code is represented in the output as a series of path lengths - * from which the canonical Huffman code can be reconstructed. The path lengths - * themselves are compressed using a separate Huffman code, the precode, which - * consists of LZX_PRECODE_NUM_SYMBOLS (= 20) symbols that cover all possible - * code lengths, plus extra codes for repeated lengths. The path lengths of the - * precode precede the path lengths of the larger code and are uncompressed, - * consisting of 20 entries of 4 bits each. - * - * @out: Bitstream to write the code to. - * @lens: The code lengths for the Huffman code, indexed by symbol. - * @prev_lens: Code lengths for this Huffman code, indexed by symbol, - * in the *previous block*, or all zeroes if this is the - * first block. - * @num_syms: The number of symbols in the code. + * Output a Huffman code in the compressed form used in LZX. + * + * The Huffman code is represented in the output as a logical series of codeword + * lengths from which the Huffman code, which must be in canonical form, can be + * reconstructed. + * + * The codeword lengths are themselves compressed using a separate Huffman code, + * the "precode", which contains a symbol for each possible codeword length in + * the larger code as well as several special symbols to represent repeated + * codeword lengths (a form of run-length encoding). The precode is itself + * constructed in canonical form, and its codeword lengths are represented + * literally in 20 4-bit fields that immediately precede the compressed codeword + * lengths of the larger code. + * + * Furthermore, the codeword lengths of the larger code are actually represented + * as deltas from the codeword lengths of the corresponding code in the previous + * block. + * + * @out: + * Bitstream to which to write the compressed Huffman code. + * @lens: + * The codeword lengths, indexed by symbol, in the Huffman code. + * @prev_lens: + * The codeword lengths, indexed by symbol, in the corresponding Huffman + * code in the previous block, or all zeroes if this is the first block. + * @num_syms: + * The number of symbols in the Huffman code. */ static void lzx_write_compressed_code(struct output_bitstream *out, @@ -837,7 +689,7 @@ lzx_write_compressed_code(struct output_bitstream *out, const u8 prev_lens[restrict], unsigned num_syms) { - freq_t precode_freqs[LZX_PRECODE_NUM_SYMBOLS]; + input_idx_t precode_freqs[LZX_PRECODE_NUM_SYMBOLS]; u8 output_syms[num_syms]; u8 precode_lens[LZX_PRECODE_NUM_SYMBOLS]; u16 precode_codewords[LZX_PRECODE_NUM_SYMBOLS]; @@ -888,20 +740,22 @@ lzx_write_compressed_code(struct output_bitstream *out, } /* - * Writes all compressed matches and literal bytes in an LZX block to the the - * output bitstream. + * Write all matches and literal bytes (which were precomputed) in an LZX + * compressed block to the output bitstream in the final compressed + * representation. * * @ostream * The output bitstream. * @block_type - * The type of the block (LZX_BLOCKTYPE_ALIGNED or LZX_BLOCKTYPE_VERBATIM). + * The chosen type of the LZX compressed block (LZX_BLOCKTYPE_ALIGNED or + * LZX_BLOCKTYPE_VERBATIM). * @match_tab - * The array of matches/literals that will be output (length @match_count). + * The array of matches/literals to output. * @match_count - * Number of matches/literals to be output. + * Number of matches/literals to output (length of @match_tab). * @codes - * Pointer to a structure that contains the codewords for the main, length, - * and aligned offset Huffman codes. + * The main, length, and aligned offset Huffman codes for the current + * LZX compressed block. */ static void lzx_write_matches_and_literals(struct output_bitstream *ostream, @@ -913,28 +767,23 @@ lzx_write_matches_and_literals(struct output_bitstream *ostream, for (unsigned i = 0; i < match_count; i++) { struct lzx_match match = match_tab[i]; - /* High bit of the match indicates whether the match is an - * actual match (1) or a literal uncompressed byte (0) */ - if (match.data & 0x80000000) { - /* match */ - lzx_write_match(ostream, block_type, - match, codes); - } else { - /* literal byte */ - bitstream_put_bits(ostream, - codes->codewords.main[match.data], - codes->lens.main[match.data]); - } + /* The high bit of the 32-bit intermediate representation + * indicates whether the item is an actual LZ-style match (1) or + * a literal byte (0). */ + if (match.data & 0x80000000) + lzx_write_match(ostream, block_type, match, codes); + else + lzx_write_literal(ostream, match.data, codes); } } static void -lzx_assert_codes_valid(const struct lzx_codes * codes) +lzx_assert_codes_valid(const struct lzx_codes * codes, unsigned num_main_syms) { #ifdef ENABLE_LZX_DEBUG unsigned i; - for (i = 0; i < LZX_MAINCODE_NUM_SYMBOLS; i++) + for (i = 0; i < num_main_syms; i++) LZX_ASSERT(codes->lens.main[i] <= LZX_MAX_MAIN_CODEWORD_LEN); for (i = 0; i < LZX_LENCODE_NUM_SYMBOLS; i++) @@ -945,10 +794,10 @@ lzx_assert_codes_valid(const struct lzx_codes * codes) const unsigned tablebits = 10; u16 decode_table[(1 << tablebits) + - (2 * max(LZX_MAINCODE_NUM_SYMBOLS, LZX_LENCODE_NUM_SYMBOLS))] + (2 * max(num_main_syms, LZX_LENCODE_NUM_SYMBOLS))] _aligned_attribute(DECODE_TABLE_ALIGNMENT); LZX_ASSERT(0 == make_huffman_decode_table(decode_table, - LZX_MAINCODE_NUM_SYMBOLS, + num_main_syms, min(tablebits, LZX_MAINCODE_TABLEBITS), codes->lens.main, LZX_MAX_MAIN_CODEWORD_LEN)); @@ -969,6 +818,8 @@ lzx_assert_codes_valid(const struct lzx_codes * codes) static void lzx_write_compressed_block(int block_type, unsigned block_size, + unsigned max_window_size, + unsigned num_main_syms, struct lzx_match * chosen_matches, unsigned num_chosen_matches, const struct lzx_codes * codes, @@ -979,27 +830,41 @@ lzx_write_compressed_block(int block_type, LZX_ASSERT(block_type == LZX_BLOCKTYPE_ALIGNED || block_type == LZX_BLOCKTYPE_VERBATIM); - LZX_ASSERT(block_size <= LZX_MAX_WINDOW_SIZE); - LZX_ASSERT(num_chosen_matches <= LZX_MAX_WINDOW_SIZE); - lzx_assert_codes_valid(codes); + lzx_assert_codes_valid(codes, num_main_syms); /* The first three bits indicate the type of block and are one of the * LZX_BLOCKTYPE_* constants. */ - bitstream_put_bits(ostream, block_type, LZX_BLOCKTYPE_NBITS); + bitstream_put_bits(ostream, block_type, 3); - /* The next bit indicates whether the block size is the default (32768), - * indicated by a 1 bit, or whether the block size is given by the next - * 16 bits, indicated by a 0 bit. */ + /* Output the block size. + * + * The original LZX format seemed to always encode the block size in 3 + * bytes. However, the implementation in WIMGAPI, as used in WIM files, + * uses the first bit to indicate whether the block is the default size + * (32768) or a different size given explicitly by the next 16 bits. + * + * By default, this compressor uses a window size of 32768 and therefore + * follows the WIMGAPI behavior. However, this compressor also supports + * window sizes greater than 32768 bytes, which do not appear to be + * supported by WIMGAPI. In such cases, we retain the default size bit + * to mean a size of 32768 bytes but output non-default block size in 24 + * bits rather than 16. The compatibility of this behavior is unknown + * because WIMs created with chunk size greater than 32768 can seemingly + * only be opened by wimlib anyway. */ if (block_size == LZX_DEFAULT_BLOCK_SIZE) { bitstream_put_bits(ostream, 1, 1); } else { bitstream_put_bits(ostream, 0, 1); - bitstream_put_bits(ostream, block_size, LZX_BLOCKSIZE_NBITS); + + if (max_window_size >= 65536) + bitstream_put_bits(ostream, block_size >> 16, 8); + + bitstream_put_bits(ostream, block_size, 16); } /* Write out lengths of the main code. Note that the LZX specification * incorrectly states that the aligned offset code comes after the - * length code, but in fact it is the very first tree to be written + * length code, but in fact it is the very first code to be written * (before the main code). */ if (block_type == LZX_BLOCKTYPE_ALIGNED) for (i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) @@ -1008,23 +873,23 @@ lzx_write_compressed_block(int block_type, LZX_DEBUG("Writing main code..."); - /* Write the pre-tree and lengths for the first LZX_NUM_CHARS symbols in + /* Write the precode and lengths for the first LZX_NUM_CHARS symbols in * the main code, which are the codewords for literal bytes. */ lzx_write_compressed_code(ostream, codes->lens.main, prev_codes->lens.main, LZX_NUM_CHARS); - /* Write the pre-tree and lengths for the rest of the main code, which + /* Write the precode and lengths for the rest of the main code, which * are the codewords for match headers. */ lzx_write_compressed_code(ostream, codes->lens.main + LZX_NUM_CHARS, prev_codes->lens.main + LZX_NUM_CHARS, - LZX_MAINCODE_NUM_SYMBOLS - LZX_NUM_CHARS); + num_main_syms - LZX_NUM_CHARS); LZX_DEBUG("Writing length code..."); - /* Write the pre-tree and lengths for the length code. */ + /* Write the precode and lengths for the length code. */ lzx_write_compressed_code(ostream, codes->lens.len, prev_codes->lens.len, @@ -1044,6 +909,7 @@ lzx_write_compressed_block(int block_type, static void lzx_write_all_blocks(struct lzx_compressor *ctx, struct output_bitstream *ostream) { + const struct lzx_codes *prev_codes = &ctx->zero_codes; for (unsigned i = 0; i < ctx->num_blocks; i++) { const struct lzx_block_spec *spec = &ctx->block_specs[i]; @@ -1055,11 +921,14 @@ lzx_write_all_blocks(struct lzx_compressor *ctx, struct output_bitstream *ostrea lzx_write_compressed_block(spec->block_type, spec->block_size, + ctx->max_window_size, + ctx->num_main_syms, &ctx->chosen_matches[spec->chosen_matches_start_pos], spec->num_chosen_matches, &spec->codes, prev_codes, ostream); + prev_codes = &spec->codes; } } @@ -1067,13 +936,10 @@ lzx_write_all_blocks(struct lzx_compressor *ctx, struct output_bitstream *ostrea /* Constructs an LZX match from a literal byte and updates the main code symbol * frequencies. */ static u32 -lzx_record_literal(u8 literal, void *_freqs) +lzx_tally_literal(u8 lit, struct lzx_freqs *freqs) { - struct lzx_freqs *freqs = _freqs; - - freqs->main[literal]++; - - return (u32)literal; + freqs->main[lit]++; + return (u32)lit; } /* Constructs an LZX match from an offset and a length, and updates the LRU @@ -1081,11 +947,9 @@ lzx_record_literal(u8 literal, void *_freqs) * alphabets. The return value is a 32-bit number that provides the match in an * intermediate representation documented below. */ static u32 -lzx_record_match(unsigned match_offset, unsigned match_len, - void *_freqs, void *_queue) +lzx_tally_match(unsigned match_len, unsigned match_offset, + struct lzx_freqs *freqs, struct lzx_lru_queue *queue) { - struct lzx_freqs *freqs = _freqs; - struct lzx_lru_queue *queue = _queue; unsigned position_slot; unsigned position_footer; u32 len_header; @@ -1134,34 +998,46 @@ lzx_record_match(unsigned match_offset, unsigned match_len, freqs->aligned[position_footer & 7]++; /* Pack the position slot, position footer, and match length into an - * intermediate representation. - * - * bits description - * ---- ----------------------------------------------------------- - * - * 31 1 if a match, 0 if a literal. - * - * 30-25 position slot. This can be at most 50, so it will fit in 6 - * bits. - * - * 8-24 position footer. This is the offset of the real formatted - * offset from the position base. This can be at most 17 bits - * (since lzx_extra_bits[LZX_NUM_POSITION_SLOTS - 1] is 17). - * - * 0-7 length of match, offset by 2. This can be at most - * (LZX_MAX_MATCH_LEN - 2) == 255, so it will fit in 8 bits. */ - BUILD_BUG_ON(LZX_NUM_POSITION_SLOTS > 64); - LZX_ASSERT(lzx_get_num_extra_bits(LZX_NUM_POSITION_SLOTS - 1) <= 17); - BUILD_BUG_ON(LZX_MAX_MATCH_LEN - LZX_MIN_MATCH_LEN + 1 > 256); + * intermediate representation. See `struct lzx_match' for details. + */ + LZX_ASSERT(LZX_MAX_POSITION_SLOTS <= 64); + LZX_ASSERT(lzx_get_num_extra_bits(LZX_MAX_POSITION_SLOTS - 1) <= 17); + LZX_ASSERT(LZX_MAX_MATCH_LEN - LZX_MIN_MATCH_LEN + 1 <= 256); + + LZX_ASSERT(position_slot <= (1U << (31 - 25)) - 1); + LZX_ASSERT(position_footer <= (1U << (25 - 8)) - 1); + LZX_ASSERT(adjusted_match_len <= (1U << (8 - 0)) - 1); return 0x80000000 | (position_slot << 25) | (position_footer << 8) | (adjusted_match_len); } +struct lzx_record_ctx { + struct lzx_freqs freqs; + struct lzx_lru_queue queue; + struct lzx_match *matches; +}; + +static void +lzx_record_match(unsigned len, unsigned offset, void *_ctx) +{ + struct lzx_record_ctx *ctx = _ctx; + + (ctx->matches++)->data = lzx_tally_match(len, offset, &ctx->freqs, &ctx->queue); +} + +static void +lzx_record_literal(u8 lit, void *_ctx) +{ + struct lzx_record_ctx *ctx = _ctx; + + (ctx->matches++)->data = lzx_tally_literal(lit, &ctx->freqs); +} + /* Returns the cost, in bits, to output a literal byte using the specified cost * model. */ -static sym_cost_t +static unsigned lzx_literal_cost(u8 c, const struct lzx_costs * costs) { return costs->main[c]; @@ -1172,18 +1048,18 @@ lzx_literal_cost(u8 c, const struct lzx_costs * costs) * codes, return the approximate number of bits it will take to represent this * match in the compressed output. Take into account the match offset LRU * queue and optionally update it. */ -static sym_cost_t +static unsigned lzx_match_cost(unsigned length, unsigned offset, const struct lzx_costs *costs, struct lzx_lru_queue *queue) { unsigned position_slot; unsigned len_header, main_symbol; - sym_cost_t cost = 0; + unsigned cost = 0; position_slot = lzx_get_position_slot(offset, queue); len_header = min(length - LZX_MIN_MATCH_LEN, LZX_NUM_PRIMARY_LENS); - main_symbol = (position_slot << 3) | len_header | LZX_NUM_CHARS; + main_symbol = ((position_slot << 3) | len_header) + LZX_NUM_CHARS; /* Account for main symbol. */ cost += costs->main[main_symbol]; @@ -1205,16 +1081,20 @@ lzx_match_cost(unsigned length, unsigned offset, const struct lzx_costs *costs, } -/* Very fast heuristic cost evaluation to use in the inner loop of the - * match-finder. */ -static sym_cost_t -lzx_match_cost_fast(unsigned offset, const struct lzx_lru_queue *queue) +/* Fast heuristic cost evaluation to use in the inner loop of the match-finder. + * Unlike lzx_match_cost() which does a true cost evaluation, this simply + * prioritize matches based on their offset. */ +static input_idx_t +lzx_match_cost_fast(input_idx_t length, input_idx_t offset, const void *_queue) { - for (unsigned i = 0; i < LZX_NUM_RECENT_OFFSETS; i++) + const struct lzx_lru_queue *queue = _queue; + + /* It seems well worth it to take the time to give priority to recently + * used offsets. */ + for (input_idx_t i = 0; i < LZX_NUM_RECENT_OFFSETS; i++) if (offset == queue->R[i]) return i; - BUILD_BUG_ON(LZX_MAX_WINDOW_SIZE >= (sym_cost_t)~0U); return offset; } @@ -1231,9 +1111,10 @@ static void lzx_set_costs(struct lzx_compressor * ctx, const struct lzx_lens * lens) { unsigned i; + unsigned num_main_syms = ctx->num_main_syms; /* Main code */ - for (i = 0; i < LZX_MAINCODE_NUM_SYMBOLS; i++) { + for (i = 0; i < num_main_syms; i++) { ctx->costs.main[i] = lens->main[i]; if (ctx->costs.main[i] == 0) ctx->costs.main[i] = ctx->params.alg_params.slow.main_nostat_cost; @@ -1254,244 +1135,10 @@ lzx_set_costs(struct lzx_compressor * ctx, const struct lzx_lens * lens) } } -/* Advance the suffix array match-finder to the next position. */ -static void -lzx_lz_update_salink(input_idx_t i, - const input_idx_t SA[restrict], - const input_idx_t ISA[restrict], - struct salink link[restrict]) -{ - /* r = Rank of the suffix at the current position. */ - const input_idx_t r = ISA[i]; - - /* next = rank of LOWEST ranked suffix that is ranked HIGHER than the - * current suffix AND has a LOWER position, or -1 if none exists. */ - const input_idx_t next = link[r].next; - - /* prev = rank of HIGHEST ranked suffix that is ranked LOWER than the - * current suffix AND has a LOWER position, or -1 if none exists. */ - const input_idx_t prev = link[r].prev; - - /* Link the suffix at the current position into the linked list that - * contains all suffixes in the suffix array that are appear at or - * before the current position, sorted by rank. - * - * Save the values of all fields we overwrite so that rollback is - * possible. */ - if (next != (input_idx_t)~0U) { - - link[next].prev = r; - link[next].lcpprev = link[r].lcpnext; - } - - if (prev != (input_idx_t)~0U) { - - link[prev].next = r; - link[prev].lcpnext = link[r].lcpprev; - } -} - -/* Rewind the suffix array match-finder to the specified position. - * - * This undoes a series of updates by lzx_lz_update_salink(). */ -static void -lzx_lz_rewind_matchfinder(struct lzx_compressor *ctx, - const unsigned orig_pos) -{ - LZX_DEBUG("Rewind match-finder %u => %u", ctx->match_window_pos, orig_pos); - - if (ctx->match_window_pos == orig_pos) - return; - - LZX_ASSERT(ctx->match_window_pos > orig_pos); - LZX_ASSERT(orig_pos == 0); - ctx->matches_cached = true; - ctx->cached_matches_pos = 0; - ctx->match_window_pos = orig_pos; -} - -/* - * Use the suffix array match-finder to retrieve a list of LZ matches at the - * current position. - * - * [in] @i Current position in the window. - * [in] @SA Suffix array for the window. - * [in] @ISA Inverse suffix array for the window. - * [inout] @link Suffix array links used internally by the match-finder. - * [out] @matches The (length, offset) pairs of the resulting matches will - * be written here, sorted in decreasing order by - * length. All returned lengths will be unique. - * [in] @queue Recently used match offsets, used when evaluating the - * cost of matches. - * [in] @min_match_len Minimum match length to return. - * [in] @max_matches_to_consider Maximum number of matches to consider at - * the position. - * [in] @max_matches_to_return Maximum number of matches to return. - * - * The return value is the number of matches found and written to @matches. - */ -static unsigned -lzx_lz_get_matches(const input_idx_t i, - const input_idx_t SA[const restrict], - const input_idx_t ISA[const restrict], - struct salink link[const restrict], - struct raw_match matches[const restrict], - const struct lzx_lru_queue * const restrict queue, - const unsigned min_match_len, - const uint32_t max_matches_to_consider, - const uint32_t max_matches_to_return) -{ - /* r = Rank of the suffix at the current position. */ - const input_idx_t r = ISA[i]; - - /* Prepare for searching the current position. */ - lzx_lz_update_salink(i, SA, ISA, link); - - /* L = rank of next suffix to the left; - * R = rank of next suffix to the right; - * lenL = length of match between current position and the suffix with rank L; - * lenR = length of match between current position and the suffix with rank R. - * - * This is left and right relative to the rank of the current suffix. - * Since the suffixes in the suffix array are sorted, the longest - * matches are immediately to the left and right (using the linked list - * to ignore all suffixes that occur later in the window). The match - * length decreases the farther left and right we go. We shall keep the - * length on both sides in sync in order to choose the lowest-cost match - * of each length. - */ - input_idx_t L = link[r].prev; - input_idx_t R = link[r].next; - input_idx_t lenL = link[r].lcpprev; - input_idx_t lenR = link[r].lcpnext; - - /* nmatches = number of matches found so far. */ - unsigned nmatches = 0; - - /* best_cost = cost of lowest-cost match found so far. - * - * We keep track of this so that we can ignore shorter matches that do - * not have lower costs than a longer matches already found. - */ - sym_cost_t best_cost = INFINITE_SYM_COST; - - /* count_remaining = maximum number of possible matches remaining to be - * considered. */ - uint32_t count_remaining = max_matches_to_consider; - - /* pending = match currently being considered for a specific length. */ - struct raw_match pending; - - while (lenL >= min_match_len || lenR >= min_match_len) - { - pending.len = lenL; - pending.offset = (input_idx_t)~0U; - sym_cost_t pending_cost = INFINITE_SYM_COST; - sym_cost_t cost; - - /* Extend left. */ - if (lenL >= min_match_len && lenL >= lenR) { - for (;;) { - - if (--count_remaining == 0) - goto out_save_pending; - - input_idx_t offset = i - SA[L]; - - /* Save match if it has smaller cost. */ - cost = lzx_match_cost_fast(offset, queue); - if (cost < pending_cost) { - pending.offset = offset; - pending_cost = cost; - } - - if (link[L].lcpprev < lenL) { - /* Match length decreased. */ - - lenL = link[L].lcpprev; - - /* Save the pending match unless the - * right side still may have matches of - * this length to be scanned, or if a - * previous (longer) match had lower - * cost. */ - if (pending.len > lenR) { - if (pending_cost < best_cost) { - best_cost = pending_cost; - matches[nmatches++] = pending; - if (nmatches == max_matches_to_return) - return nmatches; - } - pending.len = lenL; - pending.offset = (input_idx_t)~0U; - pending_cost = INFINITE_SYM_COST; - } - if (lenL < min_match_len || lenL < lenR) - break; - } - L = link[L].prev; - } - } - - pending.len = lenR; - - /* Extend right. */ - if (lenR >= min_match_len && lenR > lenL) { - for (;;) { - - if (--count_remaining == 0) - goto out_save_pending; - - input_idx_t offset = i - SA[R]; - - /* Save match if it has smaller cost. */ - cost = lzx_match_cost_fast(offset, queue); - if (cost < pending_cost) { - pending.offset = offset; - pending_cost = cost; - } - - if (link[R].lcpnext < lenR) { - /* Match length decreased. */ - - lenR = link[R].lcpnext; - - /* Save the pending match unless a - * previous (longer) match had lower - * cost. */ - if (pending_cost < best_cost) { - matches[nmatches++] = pending; - best_cost = pending_cost; - if (nmatches == max_matches_to_return) - return nmatches; - } - - if (lenR < min_match_len || lenR <= lenL) - break; - - pending.len = lenR; - pending.offset = (input_idx_t)~0U; - pending_cost = INFINITE_SYM_COST; - } - R = link[R].next; - } - } - } - goto out; - -out_save_pending: - if (pending.offset != (input_idx_t)~0U) - matches[nmatches++] = pending; - -out: - return nmatches; -} - - /* Tell the match-finder to skip the specified number of bytes (@n) in the * input. */ static void -lzx_lz_skip_bytes(struct lzx_compressor *ctx, unsigned n) +lzx_lz_skip_bytes(struct lzx_compressor *ctx, input_idx_t n) { LZX_ASSERT(n <= ctx->match_window_end - ctx->match_window_pos); if (ctx->matches_cached) { @@ -1503,22 +1150,23 @@ lzx_lz_skip_bytes(struct lzx_compressor *ctx, unsigned n) } else { while (n--) { ctx->cached_matches[ctx->cached_matches_pos++].len = 0; - lzx_lz_update_salink(ctx->match_window_pos++, ctx->SA, - ctx->ISA, ctx->salink); + lz_sarray_skip_position(&ctx->lz_sarray); + ctx->match_window_pos++; } + LZX_ASSERT(lz_sarray_get_pos(&ctx->lz_sarray) == ctx->match_window_pos); } } /* Retrieve a list of matches available at the next position in the input. * - * The matches are written to ctx->matches in decreasing order of length, and - * the return value is the number of matches found. */ -static unsigned + * A pointer to the matches array is written into @matches_ret, and the return + * value is the number of matches found. */ +static u32 lzx_lz_get_matches_caching(struct lzx_compressor *ctx, const struct lzx_lru_queue *queue, struct raw_match **matches_ret) { - unsigned num_matches; + u32 num_matches; struct raw_match *matches; LZX_ASSERT(ctx->match_window_pos <= ctx->match_window_end); @@ -1528,24 +1176,11 @@ lzx_lz_get_matches_caching(struct lzx_compressor *ctx, if (ctx->matches_cached) { num_matches = matches[-1].len; } else { - unsigned min_match_len = LZX_MIN_MATCH_LEN; - if (min_match_len <= 2 && !ctx->params.alg_params.slow.use_len2_matches) - min_match_len = 3; - const uint32_t max_search_depth = ctx->params.alg_params.slow.max_search_depth; - const uint32_t max_matches_per_pos = ctx->params.alg_params.slow.max_matches_per_pos; - - if (unlikely(max_search_depth == 0 || max_matches_per_pos == 0)) - num_matches = 0; - else - num_matches = lzx_lz_get_matches(ctx->match_window_pos, - ctx->SA, - ctx->ISA, - ctx->salink, - matches, - queue, - min_match_len, - max_search_depth, - max_matches_per_pos); + LZX_ASSERT(lz_sarray_get_pos(&ctx->lz_sarray) == ctx->match_window_pos); + num_matches = lz_sarray_get_matches(&ctx->lz_sarray, + matches, + lzx_match_cost_fast, + queue); matches[-1].len = num_matches; } ctx->cached_matches_pos += num_matches + 1; @@ -1555,7 +1190,7 @@ lzx_lz_get_matches_caching(struct lzx_compressor *ctx, * if it is not the whole window. */ if (ctx->match_window_end < ctx->window_size) { unsigned maxlen = ctx->match_window_end - ctx->match_window_pos; - for (unsigned i = 0; i < num_matches; i++) + for (u32 i = 0; i < num_matches; i++) if (matches[i].len > maxlen) matches[i].len = maxlen; } @@ -1567,7 +1202,7 @@ lzx_lz_get_matches_caching(struct lzx_compressor *ctx, #endif #ifdef ENABLE_LZX_DEBUG - for (unsigned i = 0; i < num_matches; i++) { + for (u32 i = 0; i < num_matches; i++) { LZX_ASSERT(matches[i].len >= LZX_MIN_MATCH_LEN); LZX_ASSERT(matches[i].len <= LZX_MAX_MATCH_LEN); LZX_ASSERT(matches[i].len <= ctx->match_window_end - ctx->match_window_pos); @@ -1583,421 +1218,113 @@ lzx_lz_get_matches_caching(struct lzx_compressor *ctx, return num_matches; } -/* - * Reverse the linked list of near-optimal matches so that they can be returned - * in forwards order. - * - * Returns the first match in the list. - */ -static struct raw_match -lzx_lz_reverse_near_optimal_match_list(struct lzx_compressor *ctx, - unsigned cur_pos) +static u32 +lzx_get_prev_literal_cost(struct lzx_compressor *ctx, + struct lzx_lru_queue *queue) { - unsigned prev_link, saved_prev_link; - unsigned prev_match_offset, saved_prev_match_offset; - - ctx->optimum_end_idx = cur_pos; - - saved_prev_link = ctx->optimum[cur_pos].prev.link; - saved_prev_match_offset = ctx->optimum[cur_pos].prev.match_offset; - - do { - prev_link = saved_prev_link; - prev_match_offset = saved_prev_match_offset; - - saved_prev_link = ctx->optimum[prev_link].prev.link; - saved_prev_match_offset = ctx->optimum[prev_link].prev.match_offset; - - ctx->optimum[prev_link].next.link = cur_pos; - ctx->optimum[prev_link].next.match_offset = prev_match_offset; - - cur_pos = prev_link; - } while (cur_pos != 0); - - ctx->optimum_cur_idx = ctx->optimum[0].next.link; - - return (struct raw_match) - { .len = ctx->optimum_cur_idx, - .offset = ctx->optimum[0].next.match_offset, - }; + return lzx_literal_cost(ctx->window[ctx->match_window_pos - 1], + &ctx->costs); } -#if 0 -static struct raw_match -lzx_lz_get_greedy_match(struct lzx_compressor * ctx) +static u32 +lzx_get_match_cost(struct lzx_compressor *ctx, + struct lzx_lru_queue *queue, + input_idx_t length, input_idx_t offset) { - struct raw_match *matches; - - if (!lzx_lz_get_matches_caching(ctx, &ctx->queue, &matches)) - return (struct raw_match) {.len = 0}; - - lzx_lz_skip_bytes(ctx, matches[0].len - 1); - return matches[0]; + return lzx_match_cost(length, offset, &ctx->costs, queue); } -#endif -#if 0 static struct raw_match -lzx_lz_get_lazy_match(struct lzx_compressor * ctx) +lzx_lz_get_near_optimal_match(struct lzx_compressor *ctx) { - unsigned num_matches; - struct raw_match *matches; - struct raw_match prev_match; - struct lzx_lru_queue queue; + return lz_get_near_optimal_match(&ctx->mc, + lzx_lz_get_matches_caching, + lzx_lz_skip_bytes, + lzx_get_prev_literal_cost, + lzx_get_match_cost, + ctx, + &ctx->queue); +} - if (ctx->optimum_cur_idx != ctx->optimum_end_idx) - goto retopt; +/* Set default symbol costs for the LZX Huffman codes. */ +static void +lzx_set_default_costs(struct lzx_costs * costs, unsigned num_main_syms) +{ + unsigned i; - /* Check for matches at first position. */ - num_matches = lzx_lz_get_matches_caching(ctx, &ctx->queue, &matches); + /* Main code (part 1): Literal symbols */ + for (i = 0; i < LZX_NUM_CHARS; i++) + costs->main[i] = 8; - /* Return literal if no matches were found. */ - if (num_matches == 0) - return (struct raw_match) { .len = 0 }; + /* Main code (part 2): Match header symbols */ + for (; i < num_main_syms; i++) + costs->main[i] = 10; - /* Immediately choose match if longer than threshold. */ - if (matches[0].len > ctx->params.alg_params.slow.num_fast_bytes) - goto savecur; + /* Length code */ + for (i = 0; i < LZX_LENCODE_NUM_SYMBOLS; i++) + costs->len[i] = 8; - ctx->optimum_cur_idx = ctx->optimum_end_idx = 0; - for (;;) { - prev_match = matches[0]; + /* Aligned offset code */ + for (i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) + costs->aligned[i] = 3; +} - /* Check for matches at next position. */ - num_matches = lzx_lz_get_matches_caching(ctx, &ctx->queue, &matches); +/* Given the frequencies of symbols in an LZX-compressed block and the + * corresponding Huffman codes, return LZX_BLOCKTYPE_ALIGNED or + * LZX_BLOCKTYPE_VERBATIM if an aligned offset or verbatim block, respectively, + * will take fewer bits to output. */ +static int +lzx_choose_verbatim_or_aligned(const struct lzx_freqs * freqs, + const struct lzx_codes * codes) +{ + unsigned aligned_cost = 0; + unsigned verbatim_cost = 0; - /* Choose previous match if there is not a match at this - * position. */ - if (num_matches == 0) - goto saveprev; + /* Verbatim blocks have a constant 3 bits per position footer. Aligned + * offset blocks have an aligned offset symbol per position footer, plus + * an extra 24 bits per block to output the lengths necessary to + * reconstruct the aligned offset code itself. */ + for (unsigned i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) { + verbatim_cost += 3 * freqs->aligned[i]; + aligned_cost += codes->lens.aligned[i] * freqs->aligned[i]; + } + aligned_cost += LZX_ALIGNEDCODE_ELEMENT_SIZE * LZX_ALIGNEDCODE_NUM_SYMBOLS; + if (aligned_cost < verbatim_cost) + return LZX_BLOCKTYPE_ALIGNED; + else + return LZX_BLOCKTYPE_VERBATIM; +} - /* Choose previous match the longest match at the next position - * is the same place, just one character shifted over. */ - if (matches[0].offset == prev_match.offset || - matches[0].len < prev_match.len) - goto saveprev; +/* Find a near-optimal sequence of matches/literals with which to output the + * specified LZX block, then set the block's type to that which has the minimum + * cost to output (either verbatim or aligned). */ +static void +lzx_optimize_block(struct lzx_compressor *ctx, struct lzx_block_spec *spec, + unsigned num_passes) +{ + const struct lzx_lru_queue orig_queue = ctx->queue; + struct lzx_freqs freqs; - struct lzx_lru_queue q1 = ctx->queue, q2 = ctx->queue; - double lazycost = lzx_literal_cost(ctx->window[ctx->match_window_pos - 2], - &ctx->costs) + - lzx_match_cost(matches[0].len, matches[0].offset, - &ctx->costs, &q1); - double greedycost = lzx_match_cost(prev_match.len, prev_match.offset, - &ctx->costs, &q2); - lazycost *= (double)prev_match.len / (1 + matches[0].len); + unsigned orig_window_pos = spec->window_pos; + unsigned orig_cached_pos = ctx->cached_matches_pos; - /* Choose previous match if greedy cost was lower. */ - if (greedycost <= lazycost) - goto saveprev; + LZX_ASSERT(ctx->match_window_pos == spec->window_pos); - /* Choose literal at the previous position. */ - ctx->optimum[ctx->optimum_end_idx++].next.link = 0; + ctx->match_window_end = spec->window_pos + spec->block_size; + spec->chosen_matches_start_pos = spec->window_pos; + LZX_ASSERT(num_passes >= 1); - /* Immediately choose match if longer than threshold. */ - if (matches[0].len > ctx->params.alg_params.slow.num_fast_bytes) - goto savecur; - } + /* The first optimal parsing pass is done using the cost model already + * set in ctx->costs. Each later pass is done using a cost model + * computed from the previous pass. */ + for (unsigned pass = 0; pass < num_passes; pass++) { -savecur: - lzx_lz_skip_bytes(ctx, 1); - prev_match = matches[0]; - -saveprev: - lzx_lz_skip_bytes(ctx, prev_match.len - 2); - ctx->optimum[ctx->optimum_end_idx].next.link = prev_match.len; - ctx->optimum[ctx->optimum_end_idx].next.match_offset = prev_match.offset; - ctx->optimum_end_idx++; -retopt: - prev_match.len = ctx->optimum[ctx->optimum_cur_idx].next.link; - prev_match.offset = ctx->optimum[ctx->optimum_cur_idx].next.match_offset; - ctx->optimum_cur_idx++; - return prev_match; -} -#endif - - -/* - * lzx_lz_get_near_optimal_match() - - * - * Choose the optimal match or literal to use at the next position in the input. - * - * Unlike a greedy parser that always takes the longest match, or even a - * parser with one match/literal look-ahead like zlib, the algorithm used here - * may look ahead many matches/literals to determine the optimal match/literal to - * output next. The motivation is that the compression ratio is improved if the - * compressor can do things like use a shorter-than-possible match in order to - * allow a longer match later, and also take into account the Huffman code cost - * model rather than simply assuming that longer is better. - * - * Still, this is not truly an optimal parser because very long matches are - * taken immediately. This is done to avoid considering many different - * alternatives that are unlikely to significantly be better. - * - * This algorithm is based on that used in 7-Zip's DEFLATE encoder. - * - * Each call to this function does one of two things: - * - * 1. Build a near-optimal sequence of matches/literals, up to some point, that - * will be returned by subsequent calls to this function, then return the - * first one. - * - * OR - * - * 2. Return the next match/literal previously computed by a call to this - * function; - * - * This function relies on the following state in the compressor context: - * - * ctx->window (read-only: preprocessed data being compressed) - * ctx->cost (read-only: cost model to use) - * ctx->optimum (internal state; leave uninitialized) - * ctx->optimum_cur_idx (must set to 0 before first call) - * ctx->optimum_end_idx (must set to 0 before first call) - * ctx->SA (must be built before first call) - * ctx->ISA (must be built before first call) - * ctx->salink (must be built before first call) - * ctx->match_window_pos (must initialize to position of next match to - * return; subsequent calls return subsequent - * matches) - * ctx->match_window_end (must initialize to limit of match-finding region; - * subsequent calls use the same limit) - * - * The return value is a (length, offset) pair specifying the match or literal - * chosen. For literals, the length is less than LZX_MIN_MATCH_LEN and the - * offset is meaningless. - */ -static struct raw_match -lzx_lz_get_near_optimal_match(struct lzx_compressor * ctx) -{ - unsigned num_possible_matches; - struct raw_match *possible_matches; - struct raw_match match; - unsigned longest_match_len; - - if (ctx->optimum_cur_idx != ctx->optimum_end_idx) { - /* Case 2: Return the next match/literal already found. */ - match.len = ctx->optimum[ctx->optimum_cur_idx].next.link - - ctx->optimum_cur_idx; - match.offset = ctx->optimum[ctx->optimum_cur_idx].next.match_offset; - - ctx->optimum_cur_idx = ctx->optimum[ctx->optimum_cur_idx].next.link; - return match; - } - - /* Case 1: Compute a new list of matches/literals to return. */ - - ctx->optimum_cur_idx = 0; - ctx->optimum_end_idx = 0; - - /* Get matches at this position. */ - num_possible_matches = lzx_lz_get_matches_caching(ctx, &ctx->queue, &possible_matches); - - /* If no matches found, return literal. */ - if (num_possible_matches == 0) - return (struct raw_match){ .len = 0 }; - - /* The matches that were found are sorted in decreasing order by length. - * Get the length of the longest one. */ - longest_match_len = possible_matches[0].len; - - /* Greedy heuristic: if the longest match that was found is greater - * than the number of fast bytes, return it immediately; don't both - * doing more work. */ - if (longest_match_len > ctx->params.alg_params.slow.num_fast_bytes) { - lzx_lz_skip_bytes(ctx, longest_match_len - 1); - return possible_matches[0]; - } - - /* Calculate the cost to reach the next position by outputting a - * literal. */ - ctx->optimum[0].queue = ctx->queue; - ctx->optimum[1].queue = ctx->optimum[0].queue; - ctx->optimum[1].cost = lzx_literal_cost(ctx->window[ctx->match_window_pos], - &ctx->costs); - ctx->optimum[1].prev.link = 0; - - /* Calculate the cost to reach any position up to and including that - * reached by the longest match, using the shortest (i.e. closest) match - * that reaches each position. */ - BUILD_BUG_ON(LZX_MIN_MATCH_LEN != 2); - for (unsigned len = LZX_MIN_MATCH_LEN, match_idx = num_possible_matches - 1; - len <= longest_match_len; len++) { - - LZX_ASSERT(match_idx < num_possible_matches); - - ctx->optimum[len].queue = ctx->optimum[0].queue; - ctx->optimum[len].prev.link = 0; - ctx->optimum[len].prev.match_offset = possible_matches[match_idx].offset; - ctx->optimum[len].cost = lzx_match_cost(len, - possible_matches[match_idx].offset, - &ctx->costs, - &ctx->optimum[len].queue); - if (len == possible_matches[match_idx].len) - match_idx--; - } - - unsigned cur_pos = 0; - - /* len_end: greatest index forward at which costs have been calculated - * so far */ - unsigned len_end = longest_match_len; - - for (;;) { - /* Advance to next position. */ - cur_pos++; - - if (cur_pos == len_end || cur_pos == LZX_OPTIM_ARRAY_SIZE) - return lzx_lz_reverse_near_optimal_match_list(ctx, cur_pos); - - /* retrieve the number of matches available at this position */ - num_possible_matches = lzx_lz_get_matches_caching(ctx, &ctx->optimum[cur_pos].queue, - &possible_matches); - - unsigned new_len = 0; - - if (num_possible_matches != 0) { - new_len = possible_matches[0].len; - - /* Greedy heuristic: if we found a match greater than - * the number of fast bytes, stop immediately. */ - if (new_len > ctx->params.alg_params.slow.num_fast_bytes) { - - /* Build the list of matches to return and get - * the first one. */ - match = lzx_lz_reverse_near_optimal_match_list(ctx, cur_pos); - - /* Append the long match to the end of the list. */ - ctx->optimum[cur_pos].next.match_offset = - possible_matches[0].offset; - ctx->optimum[cur_pos].next.link = cur_pos + new_len; - ctx->optimum_end_idx = cur_pos + new_len; - - /* Skip over the remaining bytes of the long match. */ - lzx_lz_skip_bytes(ctx, new_len - 1); - - /* Return first match in the list */ - return match; - } - } - - /* Consider proceeding with a literal byte. */ - block_cost_t cur_cost = ctx->optimum[cur_pos].cost; - block_cost_t cur_plus_literal_cost = cur_cost + - lzx_literal_cost(ctx->window[ctx->match_window_pos - 1], - &ctx->costs); - if (cur_plus_literal_cost < ctx->optimum[cur_pos + 1].cost) { - ctx->optimum[cur_pos + 1].cost = cur_plus_literal_cost; - ctx->optimum[cur_pos + 1].prev.link = cur_pos; - ctx->optimum[cur_pos + 1].queue = ctx->optimum[cur_pos].queue; - } - - if (num_possible_matches == 0) - continue; - - /* Consider proceeding with a match. */ - - while (len_end < cur_pos + new_len) - ctx->optimum[++len_end].cost = INFINITE_BLOCK_COST; - - for (unsigned len = LZX_MIN_MATCH_LEN, match_idx = num_possible_matches - 1; - len <= new_len; len++) { - LZX_ASSERT(match_idx < num_possible_matches); - struct lzx_lru_queue q = ctx->optimum[cur_pos].queue; - block_cost_t cost = cur_cost + lzx_match_cost(len, - possible_matches[match_idx].offset, - &ctx->costs, - &q); - - if (cost < ctx->optimum[cur_pos + len].cost) { - ctx->optimum[cur_pos + len].cost = cost; - ctx->optimum[cur_pos + len].prev.link = cur_pos; - ctx->optimum[cur_pos + len].prev.match_offset = - possible_matches[match_idx].offset; - ctx->optimum[cur_pos + len].queue = q; - } - - if (len == possible_matches[match_idx].len) - match_idx--; - } - } -} - -/* - * Set default symbol costs. - */ -static void -lzx_set_default_costs(struct lzx_costs * costs) -{ - unsigned i; - - /* Literal symbols */ - for (i = 0; i < LZX_NUM_CHARS; i++) - costs->main[i] = 8; - - /* Match header symbols */ - for (; i < LZX_MAINCODE_NUM_SYMBOLS; i++) - costs->main[i] = 10; - - /* Length symbols */ - for (i = 0; i < LZX_LENCODE_NUM_SYMBOLS; i++) - costs->len[i] = 8; - - /* Aligned offset symbols */ - for (i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) - costs->aligned[i] = 3; -} - -/* Given the frequencies of symbols in a compressed block and the corresponding - * Huffman codes, return LZX_BLOCKTYPE_ALIGNED or LZX_BLOCKTYPE_VERBATIM if an - * aligned offset or verbatim block, respectively, will take fewer bits to - * output. */ -static int -lzx_choose_verbatim_or_aligned(const struct lzx_freqs * freqs, - const struct lzx_codes * codes) -{ - unsigned aligned_cost = 0; - unsigned verbatim_cost = 0; - - /* Verbatim blocks have a constant 3 bits per position footer. Aligned - * offset blocks have an aligned offset symbol per position footer, plus - * an extra 24 bits to output the lengths necessary to reconstruct the - * aligned offset code itself. */ - for (unsigned i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) { - verbatim_cost += 3 * freqs->aligned[i]; - aligned_cost += codes->lens.aligned[i] * freqs->aligned[i]; - } - aligned_cost += LZX_ALIGNEDCODE_ELEMENT_SIZE * LZX_ALIGNEDCODE_NUM_SYMBOLS; - if (aligned_cost < verbatim_cost) - return LZX_BLOCKTYPE_ALIGNED; - else - return LZX_BLOCKTYPE_VERBATIM; -} - -/* Find a near-optimal sequence of matches/literals with which to output the - * specified LZX block, and set its type to that which has the minimum cost to - * output. */ -static void -lzx_optimize_block(struct lzx_compressor *ctx, struct lzx_block_spec *spec, - unsigned num_passes) -{ - struct lzx_lru_queue orig_queue = ctx->queue; - struct lzx_freqs freqs; - - ctx->match_window_end = spec->window_pos + spec->block_size; - spec->chosen_matches_start_pos = spec->window_pos; - - LZX_ASSERT(num_passes >= 1); - - /* The first optimal parsing pass is done using the cost model already - * set in ctx->costs. Each later pass is done using a cost model - * computed from the previous pass. */ - for (unsigned pass = 0; pass < num_passes; pass++) { - - lzx_lz_rewind_matchfinder(ctx, spec->window_pos); - ctx->queue = orig_queue; - spec->num_chosen_matches = 0; - memset(&freqs, 0, sizeof(freqs)); + ctx->match_window_pos = orig_window_pos; + ctx->cached_matches_pos = orig_cached_pos; + ctx->queue = orig_queue; + spec->num_chosen_matches = 0; + memset(&freqs, 0, sizeof(freqs)); for (unsigned i = spec->window_pos; i < spec->window_pos + spec->block_size; ) { struct raw_match raw_match; @@ -2005,30 +1332,67 @@ lzx_optimize_block(struct lzx_compressor *ctx, struct lzx_block_spec *spec, raw_match = lzx_lz_get_near_optimal_match(ctx); if (raw_match.len >= LZX_MIN_MATCH_LEN) { - lzx_match.data = lzx_record_match(raw_match.offset, raw_match.len, - &freqs, &ctx->queue); - i += raw_match.len; + if (unlikely(raw_match.len == LZX_MIN_MATCH_LEN && + raw_match.offset == ctx->max_window_size - + LZX_MIN_MATCH_LEN)) + { + /* Degenerate case where the parser + * generated the minimum match length + * with the maximum offset. There + * aren't actually enough position slots + * to represent this offset, as noted in + * the comments in + * lzx_get_num_main_syms(), so we cannot + * allow it. Use literals instead. + * + * Note that this case only occurs if + * the match-finder can generate matches + * to the very start of the window. The + * suffix array match-finder can, + * although typical hash chain and + * binary tree match-finders use 0 as a + * null value and therefore cannot + * generate such matches. */ + BUILD_BUG_ON(LZX_MIN_MATCH_LEN != 2); + lzx_match.data = lzx_tally_literal(ctx->window[i], + &freqs); + i += 1; + ctx->chosen_matches[spec->chosen_matches_start_pos + + spec->num_chosen_matches++] + = lzx_match; + lzx_match.data = lzx_tally_literal(ctx->window[i], + &freqs); + i += 1; + } else { + lzx_match.data = lzx_tally_match(raw_match.len, + raw_match.offset, + &freqs, + &ctx->queue); + i += raw_match.len; + } } else { - lzx_match.data = lzx_record_literal(ctx->window[i], &freqs); + lzx_match.data = lzx_tally_literal(ctx->window[i], &freqs); i += 1; } ctx->chosen_matches[spec->chosen_matches_start_pos + spec->num_chosen_matches++] = lzx_match; } - lzx_make_huffman_codes(&freqs, &spec->codes); + lzx_make_huffman_codes(&freqs, &spec->codes, + ctx->num_main_syms); if (pass < num_passes - 1) lzx_set_costs(ctx, &spec->codes.lens); + ctx->matches_cached = true; } spec->block_type = lzx_choose_verbatim_or_aligned(&freqs, &spec->codes); + ctx->matches_cached = false; } static void lzx_optimize_blocks(struct lzx_compressor *ctx) { lzx_lru_queue_init(&ctx->queue); - ctx->optimum_cur_idx = 0; - ctx->optimum_end_idx = 0; + lz_match_chooser_begin(&ctx->mc); const unsigned num_passes = ctx->params.alg_params.slow.num_optim_passes; @@ -2036,473 +1400,31 @@ lzx_optimize_blocks(struct lzx_compressor *ctx) lzx_optimize_block(ctx, &ctx->block_specs[i], num_passes); } -static bool entropy_val_tab_inited = false; -static double entropy_val_tab[LZX_MAX_WINDOW_SIZE]; -static pthread_mutex_t entropy_val_tab_mutex = PTHREAD_MUTEX_INITIALIZER; - -static double entropy_val(unsigned count) -{ - /*return count * log(count);*/ - return entropy_val_tab[count]; -} - -/* Split a LZX block into several if it is advantageous to do so. - * - * TODO: This doesn't work very well yet. Should optimal parsing be done - * before or after splitting? */ -static void -lzx_block_split(const u32 matches[restrict], - const input_idx_t n, - const double epsilon, - const unsigned max_num_blocks, - const unsigned min_block_len, - struct lzx_block_spec block_specs[restrict], - unsigned * const restrict num_blocks_ret) -{ - const double block_overhead = 1500; - - if (!entropy_val_tab_inited) { - pthread_mutex_lock(&entropy_val_tab_mutex); - if (!entropy_val_tab_inited) { - entropy_val_tab[0] = 0; - for (input_idx_t i = 1; i < LZX_MAX_WINDOW_SIZE; i++) - entropy_val_tab[i] = i * log2(i); - entropy_val_tab_inited = true; - } - pthread_mutex_unlock(&entropy_val_tab_mutex); - } - - u16 main_syms[n]; - u8 len_syms[n]; - u8 aligned_syms[n]; - input_idx_t orig_input_indices[n + 1]; - - LZX_ASSERT(epsilon >= 0); - LZX_ASSERT(max_num_blocks >= 1); - - /* For convenience, extract the main, length, and aligned symbols from - * the matches. Every position will have a main symbol, but not every - * position will have a length and aligned symbol. Special values - * larger than the valid symbols are used to indicate the absense of a - * symbol. */ - orig_input_indices[0] = 0; - for (input_idx_t i = 0, orig_input_idx = 0; i < n; i++) { - u32 match = matches[i]; - u16 main_sym; - u8 len_sym = LZX_LENCODE_NUM_SYMBOLS; - u8 aligned_sym = LZX_ALIGNEDCODE_NUM_SYMBOLS; - if (match & 0x80000000) { - unsigned match_len_minus_2 = match & 0xff; - unsigned position_footer = (match >> 8) & 0x1ffff; - unsigned position_slot = (match >> 25) & 0x3f; - unsigned len_header; - - if (match_len_minus_2 < LZX_NUM_PRIMARY_LENS) { - len_header = match_len_minus_2; - } else { - len_header = LZX_NUM_PRIMARY_LENS; - len_sym = match_len_minus_2 - LZX_NUM_PRIMARY_LENS; - } - main_sym = ((position_slot << 3) | len_header) + LZX_NUM_CHARS; - if (position_slot >= 8) - aligned_sym = position_footer & 7; - orig_input_idx += match_len_minus_2 + 2; - } else { - main_sym = match; - orig_input_idx++; - } - main_syms[i] = main_sym; - len_syms[i] = len_sym; - aligned_syms[i] = aligned_sym; - orig_input_indices[i + 1] = orig_input_idx; - } - - /* Compute the number of sliding windows that will be used for the - * entropy calculations. */ - int num_windows = 0; - unsigned window_len; - { - double e = min_block_len; - do { - window_len = e; - num_windows++; - e *= epsilon + 1; - } while (window_len < n); - } - - /* Compute the length of each sliding window. */ - unsigned window_lens[num_windows]; - { - double e = min_block_len; - unsigned window_idx = 0; - do { - window_len = e; - window_lens[window_idx++] = min(window_len, n); - e *= epsilon + 1; - } while (window_len < n); - } - - /* Best estimated compression size, in bits, found so far for the input - * matches up to each position. */ - unsigned shortest_paths[n + 1]; - - /* Pointers to follow to get the sequence of blocks that represents the - * shortest path (in terms of estimated compressed size) up to each - * position in the input matches. */ - input_idx_t back_ptrs[n + 1]; - - for (input_idx_t i = 0; i < n + 1; i++) { - shortest_paths[i] = ~0U; - back_ptrs[i] = 0; - } - shortest_paths[0] = 0; - - { - /* Initialize the per-window symbol and entropy counters */ - input_idx_t mainsym_ctrs[num_windows][LZX_MAINCODE_NUM_SYMBOLS]; - input_idx_t lensym_ctrs[num_windows][LZX_LENCODE_NUM_SYMBOLS + 1]; - input_idx_t alignedsym_ctrs[num_windows][LZX_ALIGNEDCODE_NUM_SYMBOLS + 1]; - ZERO_ARRAY(mainsym_ctrs); - ZERO_ARRAY(lensym_ctrs); - ZERO_ARRAY(alignedsym_ctrs); - - { - int start_win_idx = 0; - for (input_idx_t i = 0; i < n; i++) { - while (i >= window_lens[start_win_idx]) - start_win_idx++; - for (int j = start_win_idx; j < num_windows; j++) { - mainsym_ctrs[j][main_syms[i]]++; - lensym_ctrs[j][len_syms[i]]++; - alignedsym_ctrs[j][aligned_syms[i]]++; - } - } - } - - double entropy_ctrs[num_windows]; - for (int i = 0; i < num_windows; i++) { - entropy_ctrs[i] = 0; - for (unsigned j = 0; j < LZX_MAINCODE_NUM_SYMBOLS; j++) - entropy_ctrs[i] += entropy_val(mainsym_ctrs[i][j]); - for (unsigned j = 0; j < LZX_LENCODE_NUM_SYMBOLS; j++) - entropy_ctrs[i] += entropy_val(lensym_ctrs[i][j]); - for (unsigned j = 0; j < LZX_ALIGNEDCODE_NUM_SYMBOLS; j++) - entropy_ctrs[i] += entropy_val(alignedsym_ctrs[i][j]); - } - - /* Slide the windows along the input and compute the shortest - * path to each position in the matches. */ - int end_window_idx = (int)num_windows - 1; - for (input_idx_t i = 0; i < n; i++) { - for (int j = 0; j <= end_window_idx; j++) { - if (shortest_paths[i] == ~0U) - continue; - unsigned num_mainsyms = window_lens[j]; - unsigned num_lensyms = window_lens[j] - - lensym_ctrs[j][LZX_LENCODE_NUM_SYMBOLS]; - unsigned num_alignedsyms = window_lens[j] - - alignedsym_ctrs[j][LZX_ALIGNEDCODE_NUM_SYMBOLS]; - unsigned entropy = entropy_val(num_mainsyms) + - entropy_val(num_lensyms) + - entropy_val(num_alignedsyms) - - entropy_ctrs[j]; - unsigned est_csize = entropy + block_overhead; - - unsigned end_idx = i + window_lens[j]; - if (est_csize + shortest_paths[i] < shortest_paths[end_idx]) { - shortest_paths[end_idx] = est_csize + shortest_paths[i]; - back_ptrs[end_idx] = i; - } - } - /* Remove left symbol from windows */ - for (int j = 0; j <= end_window_idx; j++) { - input_idx_t orig_maincnt = mainsym_ctrs[j][main_syms[i]]--; - entropy_ctrs[j] -= entropy_val(orig_maincnt); - entropy_ctrs[j] += entropy_val(orig_maincnt - 1); - - input_idx_t orig_lencnt = - lensym_ctrs[j][len_syms[i]]--; - if (len_syms[i] != LZX_LENCODE_NUM_SYMBOLS) { - entropy_ctrs[j] -= entropy_val(orig_lencnt); - entropy_ctrs[j] += entropy_val(orig_lencnt - 1); - } - - input_idx_t orig_alignedcnt = - alignedsym_ctrs[j][aligned_syms[i]]--; - if (aligned_syms[i] != LZX_ALIGNEDCODE_NUM_SYMBOLS) { - entropy_ctrs[j] -= entropy_val(orig_alignedcnt); - entropy_ctrs[j] += entropy_val(orig_alignedcnt - 1); - } - } - - /* Calculate index of longest window remaining */ - while (end_window_idx >= 0 && window_lens[end_window_idx] >= n - i) - end_window_idx--; - - /* Append right symbol to windows */ - for (int j = 0; j <= end_window_idx; j++) { - input_idx_t orig_maincnt = mainsym_ctrs[j][ - main_syms[i + window_lens[j]]]++; - entropy_ctrs[j] -= entropy_val(orig_maincnt); - entropy_ctrs[j] += entropy_val(orig_maincnt + 1); - - input_idx_t orig_lencnt = - lensym_ctrs[j][len_syms[i + window_lens[j]]]++; - if (len_syms[i + window_lens[j]] != LZX_LENCODE_NUM_SYMBOLS) { - entropy_ctrs[j] -= entropy_val(orig_lencnt); - entropy_ctrs[j] += entropy_val(orig_lencnt + 1); - } - - input_idx_t orig_alignedcnt = - alignedsym_ctrs[j][aligned_syms[i + window_lens[j]]]++; - if (aligned_syms[i + window_lens[j]] != LZX_ALIGNEDCODE_NUM_SYMBOLS) { - entropy_ctrs[j] -= entropy_val(orig_alignedcnt); - entropy_ctrs[j] += entropy_val(orig_alignedcnt + 1); - } - } - } - } - -#if 0 - /* If no cost was computed for the first block (due to it being shorter - * than all the windows), merge it with the second block. */ - for (input_idx_t i = n; i != 0; i = back_ptrs[i]) - if (back_ptrs[i] != 0 && shortest_paths[back_ptrs[i]] == ~0U) - back_ptrs[i] = 0; -#endif - - /* Calculate number of blocks */ - input_idx_t num_blocks = 0; - for (input_idx_t i = n; i != 0; i = back_ptrs[i]) - num_blocks++; - - while (num_blocks > max_num_blocks) { - LZX_DEBUG("Joining blocks to bring total under max_num_blucks=%u", - max_num_blocks); - back_ptrs[n] = back_ptrs[back_ptrs[n]]; - num_blocks--; - } - - LZX_ASSERT(num_blocks != 0); - - /* fill in the 'struct lzx_block_spec' for each block */ - for (input_idx_t i = n, j = num_blocks - 1; i != 0; i = back_ptrs[i], j--) { - - block_specs[j].chosen_matches_start_pos = back_ptrs[i]; - block_specs[j].num_chosen_matches = i - back_ptrs[i]; - block_specs[j].window_pos = orig_input_indices[back_ptrs[i]]; - block_specs[j].block_size = orig_input_indices[i] - - orig_input_indices[back_ptrs[i]]; - /*block_specs[j].est_csize = (shortest_paths[i] -*/ - /*shortest_paths[back_ptrs[i]]) / 8;*/ - - LZX_DEBUG("block match_indices [%u, %u) est_csize %u bits\n", - back_ptrs[i], i, - shortest_paths[i] - shortest_paths[back_ptrs[i]]); - - struct lzx_freqs freqs = {}; - - for (input_idx_t k = back_ptrs[i]; k < i; k++) { - freqs.main[main_syms[k]]++; - if (len_syms[k] != LZX_LENCODE_NUM_SYMBOLS) - freqs.len[len_syms[k]]++; - if (aligned_syms[k] != LZX_LENCODE_NUM_SYMBOLS) - freqs.aligned[aligned_syms[k]]++; - } - lzx_make_huffman_codes(&freqs, &block_specs[j].codes); - - block_specs[j].block_type = lzx_choose_verbatim_or_aligned(&freqs, - &block_specs[j].codes); - } - *num_blocks_ret = num_blocks; -} - - -/* Initialize the suffix array match-finder for the specified input. */ -static void -lzx_lz_init_matchfinder(const u8 T[const restrict], - const input_idx_t n, - input_idx_t SA[const restrict], - input_idx_t ISA[const restrict], - input_idx_t LCP[const restrict], - struct salink link[const restrict], - const unsigned max_match_len) -{ - /* Compute SA (Suffix Array). */ - - { - saidx_t sa[n]; - /* ISA and link are used as temporary space. */ - BUILD_BUG_ON(LZX_MAX_WINDOW_SIZE * sizeof(ISA[0]) < 256 * sizeof(saidx_t)); - BUILD_BUG_ON(LZX_MAX_WINDOW_SIZE * 2 * sizeof(link[0]) < 256 * 256 * sizeof(saidx_t)); - divsufsort(T, sa, n, (saidx_t*)ISA, (saidx_t*)link); - for (input_idx_t i = 0; i < n; i++) - SA[i] = sa[i]; - } - -#ifdef ENABLE_LZX_DEBUG - - LZX_ASSERT(n > 0); - - /* Verify suffix array. */ - { - bool found[n]; - ZERO_ARRAY(found); - for (input_idx_t r = 0; r < n; r++) { - input_idx_t i = SA[r]; - LZX_ASSERT(i < n); - LZX_ASSERT(!found[i]); - found[i] = true; - } - } - - for (input_idx_t r = 0; r < n - 1; r++) { - - input_idx_t i1 = SA[r]; - input_idx_t i2 = SA[r + 1]; - - input_idx_t n1 = n - i1; - input_idx_t n2 = n - i2; - - LZX_ASSERT(memcmp(&T[i1], &T[i2], min(n1, n2)) <= 0); - } - LZX_DEBUG("Verified SA (len %u)", n); -#endif /* ENABLE_LZX_DEBUG */ - - /* Compute ISA (Inverse Suffix Array) */ - for (input_idx_t r = 0; r < n; r++) - ISA[SA[r]] = r; - - /* Compute LCP (longest common prefix) array. - * - * Algorithm adapted from Kasai et al. 2001: "Linear-Time - * Longest-Common-Prefix Computation in Suffix Arrays and Its - * Applications". */ - { - input_idx_t h = 0; - for (input_idx_t i = 0; i < n; i++) { - input_idx_t r = ISA[i]; - if (r > 0) { - input_idx_t j = SA[r - 1]; - - input_idx_t lim = min(n - i, n - j); - - while (h < lim && T[i + h] == T[j + h]) - h++; - LCP[r] = h; - if (h > 0) - h--; - } - } - } - -#ifdef ENABLE_LZX_DEBUG - /* Verify LCP array. */ - for (input_idx_t r = 0; r < n - 1; r++) { - LZX_ASSERT(ISA[SA[r]] == r); - LZX_ASSERT(ISA[SA[r + 1]] == r + 1); - - input_idx_t i1 = SA[r]; - input_idx_t i2 = SA[r + 1]; - input_idx_t lcp = LCP[r + 1]; - - input_idx_t n1 = n - i1; - input_idx_t n2 = n - i2; - - LZX_ASSERT(lcp <= min(n1, n2)); - - LZX_ASSERT(memcmp(&T[i1], &T[i2], lcp) == 0); - if (lcp < min(n1, n2)) - LZX_ASSERT(T[i1 + lcp] != T[i2 + lcp]); - } -#endif /* ENABLE_LZX_DEBUG */ - - /* Compute salink.next and salink.lcpnext. - * - * Algorithm adapted from Crochemore et al. 2009: - * "LPF computation revisited". - * - * Note: we cap lcpnext to the maximum match length so that the - * match-finder need not worry about it later. */ - link[n - 1].next = (input_idx_t)~0U; - link[n - 1].prev = (input_idx_t)~0U; - link[n - 1].lcpnext = 0; - link[n - 1].lcpprev = 0; - for (input_idx_t r = n - 2; r != (input_idx_t)~0U; r--) { - input_idx_t t = r + 1; - input_idx_t l = LCP[t]; - while (t != (input_idx_t)~0 && SA[t] > SA[r]) { - l = min(l, link[t].lcpnext); - t = link[t].next; - } - link[r].next = t; - link[r].lcpnext = min(l, max_match_len); - LZX_ASSERT(t == (input_idx_t)~0 || l <= n - SA[t]); - LZX_ASSERT(l <= n - SA[r]); - LZX_ASSERT(memcmp(&T[SA[r]], &T[SA[t]], l) == 0); - } - - /* Compute salink.prev and salink.lcpprev. - * - * Algorithm adapted from Crochemore et al. 2009: - * "LPF computation revisited". - * - * Note: we cap lcpprev to the maximum match length so that the - * match-finder need not worry about it later. */ - link[0].prev = (input_idx_t)~0; - link[0].next = (input_idx_t)~0; - link[0].lcpprev = 0; - link[0].lcpnext = 0; - for (input_idx_t r = 1; r < n; r++) { - input_idx_t t = r - 1; - input_idx_t l = LCP[r]; - while (t != (input_idx_t)~0 && SA[t] > SA[r]) { - l = min(l, link[t].lcpprev); - t = link[t].prev; - } - link[r].prev = t; - link[r].lcpprev = min(l, max_match_len); - LZX_ASSERT(t == (input_idx_t)~0 || l <= n - SA[t]); - LZX_ASSERT(l <= n - SA[r]); - LZX_ASSERT(memcmp(&T[SA[r]], &T[SA[t]], l) == 0); - } -} - /* Prepare the input window into one or more LZX blocks ready to be output. */ static void lzx_prepare_blocks(struct lzx_compressor * ctx) { /* Initialize the match-finder. */ - lzx_lz_init_matchfinder(ctx->window, ctx->window_size, - ctx->SA, ctx->ISA, ctx->LCP, ctx->salink, - LZX_MAX_MATCH_LEN); + lz_sarray_load_window(&ctx->lz_sarray, ctx->window, ctx->window_size); ctx->cached_matches_pos = 0; ctx->matches_cached = false; ctx->match_window_pos = 0; /* Set up a default cost model. */ - lzx_set_default_costs(&ctx->costs); + lzx_set_default_costs(&ctx->costs, ctx->num_main_syms); - /* Initially assume that the entire input will be one LZX block. */ - ctx->block_specs[0].block_type = LZX_BLOCKTYPE_ALIGNED; - ctx->block_specs[0].window_pos = 0; - ctx->block_specs[0].block_size = ctx->window_size; - ctx->num_blocks = 1; + /* TODO: The compression ratio could be slightly improved by performing + * data-dependent block splitting instead of using fixed-size blocks. + * Doing so well is a computationally hard problem, however. */ + ctx->num_blocks = DIV_ROUND_UP(ctx->window_size, LZX_DIV_BLOCK_SIZE); + for (unsigned i = 0; i < ctx->num_blocks; i++) { + unsigned pos = LZX_DIV_BLOCK_SIZE * i; + ctx->block_specs[i].window_pos = pos; + ctx->block_specs[i].block_size = min(ctx->window_size - pos, LZX_DIV_BLOCK_SIZE); + } - /* Perform near-optimal LZ parsing. */ + /* Determine sequence of matches/literals to output for each block. */ lzx_optimize_blocks(ctx); - - /* Possibly divide up the LZX block. */ - const unsigned max_num_blocks = 1U << ctx->params.alg_params.slow.num_split_passes; - if (max_num_blocks > 1) { - const double epsilon = 0.2; - const unsigned min_block_len = 500; - - lzx_block_split((const u32*)ctx->chosen_matches, - ctx->block_specs[0].num_chosen_matches, - epsilon, max_num_blocks, min_block_len, - ctx->block_specs, &ctx->num_blocks); - } } /* @@ -2516,9 +1438,6 @@ lzx_prepare_blocks(struct lzx_compressor * ctx) * ctx->window[] * ctx->window_size * - * Working space: - * ctx->queue - * * Output --- the block specification and the corresponding match/literal data: * * ctx->block_specs[] @@ -2528,8 +1447,7 @@ lzx_prepare_blocks(struct lzx_compressor * ctx) static void lzx_prepare_block_fast(struct lzx_compressor * ctx) { - unsigned num_matches; - struct lzx_freqs freqs; + struct lzx_record_ctx record_ctx; struct lzx_block_spec *spec; /* Parameters to hash chain LZ match finder @@ -2539,6 +1457,7 @@ lzx_prepare_block_fast(struct lzx_compressor * ctx) * aren't worth choosing when using greedy or lazy parsing. */ .min_match = 3, .max_match = LZX_MAX_MATCH_LEN, + .max_offset = LZX_MAX_WINDOW_SIZE, .good_match = LZX_MAX_MATCH_LEN, .nice_match = LZX_MAX_MATCH_LEN, .max_chain_len = LZX_MAX_MATCH_LEN, @@ -2547,29 +1466,28 @@ lzx_prepare_block_fast(struct lzx_compressor * ctx) }; /* Initialize symbol frequencies and match offset LRU queue. */ - memset(&freqs, 0, sizeof(struct lzx_freqs)); - lzx_lru_queue_init(&ctx->queue); + memset(&record_ctx.freqs, 0, sizeof(struct lzx_freqs)); + lzx_lru_queue_init(&record_ctx.queue); + record_ctx.matches = ctx->chosen_matches; /* Determine series of matches/literals to output. */ - num_matches = lz_analyze_block(ctx->window, - ctx->window_size, - (u32*)ctx->chosen_matches, - lzx_record_match, - lzx_record_literal, - &freqs, - &ctx->queue, - &freqs, - &lzx_lz_params); - + lz_analyze_block(ctx->window, + ctx->window_size, + lzx_record_match, + lzx_record_literal, + &record_ctx, + &lzx_lz_params, + ctx->prev_tab); /* Set up block specification. */ spec = &ctx->block_specs[0]; spec->block_type = LZX_BLOCKTYPE_ALIGNED; spec->window_pos = 0; spec->block_size = ctx->window_size; - spec->num_chosen_matches = num_matches; + spec->num_chosen_matches = (record_ctx.matches - ctx->chosen_matches); spec->chosen_matches_start_pos = 0; - lzx_make_huffman_codes(&freqs, &spec->codes); + lzx_make_huffman_codes(&record_ctx.freqs, &spec->codes, + ctx->num_main_syms); ctx->num_blocks = 1; } @@ -2607,35 +1525,32 @@ do_call_insn_preprocessing(u8 data[], int size) } } -/* API function documented in wimlib.h */ -WIMLIBAPI unsigned -wimlib_lzx_compress2(const void * const restrict uncompressed_data, - unsigned const uncompressed_len, - void * const restrict compressed_data, - struct wimlib_lzx_context * const restrict lzx_ctx) +static size_t +lzx_compress(const void *uncompressed_data, size_t uncompressed_size, + void *compressed_data, size_t compressed_size_avail, void *_ctx) { - struct lzx_compressor *ctx = (struct lzx_compressor*)lzx_ctx; + struct lzx_compressor *ctx = _ctx; struct output_bitstream ostream; - unsigned compressed_len; + size_t compressed_size; - if (uncompressed_len < 100) { + if (uncompressed_size < 100) { LZX_DEBUG("Too small to bother compressing."); return 0; } - if (uncompressed_len > 32768) { - LZX_DEBUG("Only up to 32768 bytes of uncompressed data are supported."); + if (uncompressed_size > ctx->max_window_size) { + LZX_DEBUG("Can't compress %zu bytes using window of %u bytes!", + uncompressed_size, ctx->max_window_size); return 0; } - wimlib_assert(lzx_ctx != NULL); - - LZX_DEBUG("Attempting to compress %u bytes...", uncompressed_len); + LZX_DEBUG("Attempting to compress %zu bytes...", + uncompressed_size); /* The input data must be preprocessed. To avoid changing the original * input, copy it to a temporary buffer. */ - memcpy(ctx->window, uncompressed_data, uncompressed_len); - ctx->window_size = uncompressed_len; + memcpy(ctx->window, uncompressed_data, uncompressed_size); + ctx->window_size = uncompressed_size; /* This line is unnecessary; it just avoids inconsequential accesses of * uninitialized memory that would show up in memory-checking tools such @@ -2659,323 +1574,311 @@ wimlib_lzx_compress2(const void * const restrict uncompressed_data, LZX_DEBUG("Writing compressed blocks..."); /* Generate the compressed data. */ - init_output_bitstream(&ostream, compressed_data, ctx->window_size - 1); + init_output_bitstream(&ostream, compressed_data, compressed_size_avail); lzx_write_all_blocks(ctx, &ostream); LZX_DEBUG("Flushing bitstream..."); - if (flush_output_bitstream(&ostream)) { - /* If the bitstream cannot be flushed, then the output space was - * exhausted. */ - LZX_DEBUG("Data did not compress to less than original length!"); + compressed_size = flush_output_bitstream(&ostream); + if (compressed_size == ~(input_idx_t)0) { + LZX_DEBUG("Data did not compress to %zu bytes or less!", + compressed_size_avail); return 0; } - /* Compute the length of the compressed data. */ - compressed_len = ostream.bit_output - (u8*)compressed_data; - - LZX_DEBUG("Done: compressed %u => %u bytes.", - uncompressed_len, compressed_len); + LZX_DEBUG("Done: compressed %zu => %zu bytes.", + uncompressed_size, compressed_size); /* Verify that we really get the same thing back when decompressing. - * TODO: Disable this check by default on the slow algorithm. */ + * Although this could be disabled by default in all cases, it only + * takes around 2-3% of the running time of the slow algorithm to do the + * verification. */ if (ctx->params.algorithm == WIMLIB_LZX_ALGORITHM_SLOW #if defined(ENABLE_LZX_DEBUG) || defined(ENABLE_VERIFY_COMPRESSION) || 1 #endif ) { - u8 buf[uncompressed_len]; - int ret; - - ret = wimlib_lzx_decompress(compressed_data, compressed_len, - buf, uncompressed_len); - if (ret) { - ERROR("Failed to decompress data we " - "compressed using LZX algorithm"); - wimlib_assert(0); - return 0; - } + struct wimlib_decompressor *decompressor; - if (memcmp(uncompressed_data, buf, uncompressed_len)) { - ERROR("Data we compressed using LZX algorithm " - "didn't decompress to original"); - wimlib_assert(0); - return 0; + if (0 == wimlib_create_decompressor(WIMLIB_COMPRESSION_TYPE_LZX, + ctx->max_window_size, + NULL, + &decompressor)) + { + int ret; + ret = wimlib_decompress(compressed_data, + compressed_size, + ctx->window, + uncompressed_size, + decompressor); + wimlib_free_decompressor(decompressor); + + if (ret) { + ERROR("Failed to decompress data we " + "compressed using LZX algorithm"); + wimlib_assert(0); + return 0; + } + if (memcmp(uncompressed_data, ctx->window, uncompressed_size)) { + ERROR("Data we compressed using LZX algorithm " + "didn't decompress to original"); + wimlib_assert(0); + return 0; + } + } else { + WARNING("Failed to create decompressor for " + "data verification!"); } } - return compressed_len; + return compressed_size; } -static bool -lzx_params_compatible(const struct wimlib_lzx_params *oldparams, - const struct wimlib_lzx_params *newparams) -{ - return 0 == memcmp(oldparams, newparams, sizeof(struct wimlib_lzx_params)); -} - -static struct wimlib_lzx_params lzx_user_default_params; -static struct wimlib_lzx_params *lzx_user_default_params_ptr; - -static bool -lzx_params_valid(const struct wimlib_lzx_params *params) +static void +lzx_free_compressor(void *_ctx) { - /* Validate parameters. */ - if (params->size_of_this != sizeof(struct wimlib_lzx_params)) { - LZX_DEBUG("Invalid parameter structure size!"); - return false; - } - - if (params->algorithm != WIMLIB_LZX_ALGORITHM_SLOW && - params->algorithm != WIMLIB_LZX_ALGORITHM_FAST) - { - LZX_DEBUG("Invalid algorithm."); - return false; - } - - if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) { - if (params->alg_params.slow.num_optim_passes < 1) - { - LZX_DEBUG("Invalid number of optimization passes!"); - return false; - } - - if (params->alg_params.slow.main_nostat_cost < 1 || - params->alg_params.slow.main_nostat_cost > 16) - { - LZX_DEBUG("Invalid main_nostat_cost!"); - return false; - } - - if (params->alg_params.slow.len_nostat_cost < 1 || - params->alg_params.slow.len_nostat_cost > 16) - { - LZX_DEBUG("Invalid len_nostat_cost!"); - return false; - } - - if (params->alg_params.slow.aligned_nostat_cost < 1 || - params->alg_params.slow.aligned_nostat_cost > 8) - { - LZX_DEBUG("Invalid aligned_nostat_cost!"); - return false; - } + struct lzx_compressor *ctx = _ctx; - if (params->alg_params.slow.num_split_passes > 31) { - LZX_DEBUG("Invalid num_split_passes!"); - return false; - } + if (ctx) { + FREE(ctx->chosen_matches); + FREE(ctx->cached_matches); + lz_match_chooser_destroy(&ctx->mc); + lz_sarray_destroy(&ctx->lz_sarray); + FREE(ctx->block_specs); + FREE(ctx->prev_tab); + FREE(ctx->window); + FREE(ctx); } - return true; } -WIMLIBAPI int -wimlib_lzx_set_default_params(const struct wimlib_lzx_params * params) +static const struct wimlib_lzx_compressor_params lzx_fast_default = { + .hdr = { + .size = sizeof(struct wimlib_lzx_compressor_params), + }, + .algorithm = WIMLIB_LZX_ALGORITHM_FAST, + .use_defaults = 0, + .alg_params = { + .fast = { + }, + }, +}; +static const struct wimlib_lzx_compressor_params lzx_slow_default = { + .hdr = { + .size = sizeof(struct wimlib_lzx_compressor_params), + }, + .algorithm = WIMLIB_LZX_ALGORITHM_SLOW, + .use_defaults = 0, + .alg_params = { + .slow = { + .use_len2_matches = 1, + .nice_match_length = 32, + .num_optim_passes = 2, + .max_search_depth = 50, + .max_matches_per_pos = 3, + .main_nostat_cost = 15, + .len_nostat_cost = 15, + .aligned_nostat_cost = 7, + }, + }, +}; + +static const struct wimlib_lzx_compressor_params * +lzx_get_params(const struct wimlib_compressor_params_header *_params) { - if (params) { - if (!lzx_params_valid(params)) - return WIMLIB_ERR_INVALID_PARAM; - lzx_user_default_params = *params; - lzx_user_default_params_ptr = &lzx_user_default_params; + const struct wimlib_lzx_compressor_params *params = + (const struct wimlib_lzx_compressor_params*)_params; + + if (params == NULL) { + LZX_DEBUG("Using default algorithm and parameters."); + params = &lzx_slow_default; } else { - lzx_user_default_params_ptr = NULL; + if (params->use_defaults) { + if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) + params = &lzx_slow_default; + else + params = &lzx_fast_default; + } } - return 0; + return params; } -/* API function documented in wimlib.h */ -WIMLIBAPI int -wimlib_lzx_alloc_context(const struct wimlib_lzx_params *params, - struct wimlib_lzx_context **ctx_pp) +static int +lzx_create_compressor(size_t window_size, + const struct wimlib_compressor_params_header *_params, + void **ctx_ret) { - - LZX_DEBUG("Allocating LZX context..."); - + const struct wimlib_lzx_compressor_params *params = lzx_get_params(_params); struct lzx_compressor *ctx; - static const struct wimlib_lzx_params fast_default = { - .size_of_this = sizeof(struct wimlib_lzx_params), - .algorithm = WIMLIB_LZX_ALGORITHM_FAST, - .use_defaults = 0, - .alg_params = { - .fast = { - }, - }, - }; - static const struct wimlib_lzx_params slow_default = { - .size_of_this = sizeof(struct wimlib_lzx_params), - .algorithm = WIMLIB_LZX_ALGORITHM_SLOW, - .use_defaults = 0, - .alg_params = { - .slow = { - .use_len2_matches = 1, - .num_fast_bytes = 32, - .num_optim_passes = 2, - .num_split_passes = 0, - .max_search_depth = 50, - .max_matches_per_pos = 3, - .main_nostat_cost = 15, - .len_nostat_cost = 15, - .aligned_nostat_cost = 7, - }, - }, - }; - - if (params) { - if (!lzx_params_valid(params)) - return WIMLIB_ERR_INVALID_PARAM; - } else { - LZX_DEBUG("Using default algorithm and parameters."); - if (lzx_user_default_params_ptr) - params = lzx_user_default_params_ptr; - else - params = &slow_default; - } - - if (params->use_defaults) { - if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) - params = &slow_default; - else - params = &fast_default; - } - - if (ctx_pp) { - ctx = *(struct lzx_compressor**)ctx_pp; + LZX_DEBUG("Allocating LZX context..."); - if (ctx && lzx_params_compatible(&ctx->params, params)) - return 0; - } else { - LZX_DEBUG("Check parameters only."); - return 0; - } + if (!lzx_window_size_valid(window_size)) + return WIMLIB_ERR_INVALID_PARAM; LZX_DEBUG("Allocating memory."); - ctx = MALLOC(sizeof(struct lzx_compressor)); + ctx = CALLOC(1, sizeof(struct lzx_compressor)); if (ctx == NULL) - goto err; - - size_t block_specs_length; + goto oom; + + ctx->num_main_syms = lzx_get_num_main_syms(window_size); + ctx->max_window_size = window_size; + ctx->window = MALLOC(window_size + 12); + if (ctx->window == NULL) + goto oom; + + if (params->algorithm == WIMLIB_LZX_ALGORITHM_FAST) { + ctx->prev_tab = MALLOC(window_size * sizeof(ctx->prev_tab[0])); + if (ctx->prev_tab == NULL) + goto oom; + } - if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) - block_specs_length = 1U << params->alg_params.slow.num_split_passes; - else - block_specs_length = 1U; + size_t block_specs_length = DIV_ROUND_UP(window_size, LZX_DIV_BLOCK_SIZE); ctx->block_specs = MALLOC(block_specs_length * sizeof(ctx->block_specs[0])); if (ctx->block_specs == NULL) - goto err_free_ctx; + goto oom; if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) { - ctx->SA = MALLOC(3U * LZX_MAX_WINDOW_SIZE * sizeof(ctx->SA[0])); - if (ctx->SA == NULL) - goto err_free_block_specs; - ctx->ISA = ctx->SA + LZX_MAX_WINDOW_SIZE; - ctx->LCP = ctx->ISA + LZX_MAX_WINDOW_SIZE; - ctx->salink = MALLOC(LZX_MAX_WINDOW_SIZE * sizeof(ctx->salink[0])); - if (ctx->salink == NULL) - goto err_free_SA; - } else { - ctx->SA = NULL; - ctx->ISA = NULL; - ctx->LCP = NULL; - ctx->salink = NULL; + unsigned min_match_len = LZX_MIN_MATCH_LEN; + if (!params->alg_params.slow.use_len2_matches) + min_match_len = max(min_match_len, 3); + + if (!lz_sarray_init(&ctx->lz_sarray, + window_size, + min_match_len, + LZX_MAX_MATCH_LEN, + params->alg_params.slow.max_search_depth, + params->alg_params.slow.max_matches_per_pos)) + goto oom; } if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) { - ctx->optimum = MALLOC((LZX_OPTIM_ARRAY_SIZE + LZX_MAX_MATCH_LEN) * - sizeof(ctx->optimum[0])); - if (ctx->optimum == NULL) - goto err_free_salink; - } else { - ctx->optimum = NULL; + if (!lz_match_chooser_init(&ctx->mc, + LZX_OPTIM_ARRAY_SIZE, + params->alg_params.slow.nice_match_length, + LZX_MAX_MATCH_LEN)) + goto oom; } if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) { - uint32_t cache_per_pos; + u32 cache_per_pos; cache_per_pos = params->alg_params.slow.max_matches_per_pos; if (cache_per_pos > LZX_MAX_CACHE_PER_POS) cache_per_pos = LZX_MAX_CACHE_PER_POS; - ctx->cached_matches = MALLOC(LZX_MAX_WINDOW_SIZE * (cache_per_pos + 1) * + ctx->cached_matches = MALLOC(window_size * (cache_per_pos + 1) * sizeof(ctx->cached_matches[0])); if (ctx->cached_matches == NULL) - goto err_free_optimum; - } else { - ctx->cached_matches = NULL; + goto oom; } - ctx->chosen_matches = MALLOC(LZX_MAX_WINDOW_SIZE * - sizeof(ctx->chosen_matches[0])); + ctx->chosen_matches = MALLOC(window_size * sizeof(ctx->chosen_matches[0])); if (ctx->chosen_matches == NULL) - goto err_free_cached_matches; + goto oom; - memcpy(&ctx->params, params, sizeof(struct wimlib_lzx_params)); + memcpy(&ctx->params, params, sizeof(struct wimlib_lzx_compressor_params)); memset(&ctx->zero_codes, 0, sizeof(ctx->zero_codes)); LZX_DEBUG("Successfully allocated new LZX context."); - wimlib_lzx_free_context(*ctx_pp); - *ctx_pp = (struct wimlib_lzx_context*)ctx; + *ctx_ret = ctx; return 0; -err_free_cached_matches: - FREE(ctx->cached_matches); -err_free_optimum: - FREE(ctx->optimum); -err_free_salink: - FREE(ctx->salink); -err_free_SA: - FREE(ctx->SA); -err_free_block_specs: - FREE(ctx->block_specs); -err_free_ctx: - FREE(ctx); -err: - LZX_DEBUG("Ran out of memory."); +oom: + lzx_free_compressor(ctx); return WIMLIB_ERR_NOMEM; } -/* API function documented in wimlib.h */ -WIMLIBAPI void -wimlib_lzx_free_context(struct wimlib_lzx_context *_ctx) +static u64 +lzx_get_needed_memory(size_t max_block_size, + const struct wimlib_compressor_params_header *_params) { - struct lzx_compressor *ctx = (struct lzx_compressor*)_ctx; + const struct wimlib_lzx_compressor_params *params = lzx_get_params(_params); - if (ctx) { - FREE(ctx->cached_matches); - FREE(ctx->chosen_matches); - FREE(ctx->optimum); - FREE(ctx->SA); - FREE(ctx->salink); - FREE(ctx->block_specs); - FREE(ctx); + u64 size = 0; + + size += sizeof(struct lzx_compressor); + + size += max_block_size + 12; + + size += DIV_ROUND_UP(max_block_size, LZX_DIV_BLOCK_SIZE) * + sizeof(((struct lzx_compressor*)0)->block_specs[0]); + + if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW) { + size += max_block_size * sizeof(((struct lzx_compressor*)0)->chosen_matches[0]); + size += lz_sarray_get_needed_memory(max_block_size); + size += lz_match_chooser_get_needed_memory(LZX_OPTIM_ARRAY_SIZE, + params->alg_params.slow.nice_match_length, + LZX_MAX_MATCH_LEN); + u32 cache_per_pos; + + cache_per_pos = params->alg_params.slow.max_matches_per_pos; + if (cache_per_pos > LZX_MAX_CACHE_PER_POS) + cache_per_pos = LZX_MAX_CACHE_PER_POS; + + size += max_block_size * (cache_per_pos + 1) * + sizeof(((struct lzx_compressor*)0)->cached_matches[0]); + } else { + size += max_block_size * sizeof(((struct lzx_compressor*)0)->prev_tab[0]); } + return size; } -/* API function documented in wimlib.h */ -WIMLIBAPI unsigned -wimlib_lzx_compress(const void * const restrict uncompressed_data, - unsigned const uncompressed_len, - void * const restrict compressed_data) +static bool +lzx_params_valid(const struct wimlib_compressor_params_header *_params) { - int ret; - struct wimlib_lzx_context *ctx = NULL; - unsigned compressed_len; - - ret = wimlib_lzx_alloc_context(NULL, &ctx); - if (ret) { - wimlib_assert(ret != WIMLIB_ERR_INVALID_PARAM); - WARNING("Couldn't allocate LZX compression context: %"TS"", - wimlib_get_error_string(ret)); - return 0; + const struct wimlib_lzx_compressor_params *params = + (const struct wimlib_lzx_compressor_params*)_params; + + if (params->hdr.size != sizeof(struct wimlib_lzx_compressor_params)) { + LZX_DEBUG("Invalid parameter structure size!"); + return false; } - compressed_len = wimlib_lzx_compress2(uncompressed_data, - uncompressed_len, - compressed_data, - ctx); + if (params->algorithm != WIMLIB_LZX_ALGORITHM_SLOW && + params->algorithm != WIMLIB_LZX_ALGORITHM_FAST) + { + LZX_DEBUG("Invalid algorithm."); + return false; + } + + if (params->algorithm == WIMLIB_LZX_ALGORITHM_SLOW && + !params->use_defaults) + { + if (params->alg_params.slow.num_optim_passes < 1) + { + LZX_DEBUG("Invalid number of optimization passes!"); + return false; + } + + if (params->alg_params.slow.main_nostat_cost < 1 || + params->alg_params.slow.main_nostat_cost > 16) + { + LZX_DEBUG("Invalid main_nostat_cost!"); + return false; + } - wimlib_lzx_free_context(ctx); + if (params->alg_params.slow.len_nostat_cost < 1 || + params->alg_params.slow.len_nostat_cost > 16) + { + LZX_DEBUG("Invalid len_nostat_cost!"); + return false; + } - return compressed_len; + if (params->alg_params.slow.aligned_nostat_cost < 1 || + params->alg_params.slow.aligned_nostat_cost > 8) + { + LZX_DEBUG("Invalid aligned_nostat_cost!"); + return false; + } + } + return true; } + +const struct compressor_ops lzx_compressor_ops = { + .params_valid = lzx_params_valid, + .get_needed_memory = lzx_get_needed_memory, + .create_compressor = lzx_create_compressor, + .compress = lzx_compress, + .free_compressor = lzx_free_compressor, +};