diff options
Diffstat (limited to 'lzip.c')
| -rw-r--r-- | lzip.c | 881 |
1 files changed, 881 insertions, 0 deletions
@@ -0,0 +1,881 @@ +/* + * LZIP decompressor + * + * Copyright (C) 2016-2018 Antonio Diaz Diaz. + * + * Licensed under GPLv2 or later, see file LICENSE in this source tree. + */ + +#include "linux_lzip.h" +#include "linux_mm.h" + +/* + * STATIC_RW_DATA is used in the pre-boot environment on some architectures. + * See <linux/decompress/mm.h> for details. + */ +#ifndef STATIC_RW_DATA +#define STATIC_RW_DATA static +#endif + +typedef int State; + +enum { states = 12 }; + +static inline bool St_is_char(const State st) { return st < 7; } + +static inline State St_set_char(const State st) +{ + STATIC_RW_DATA const State next[states] = { 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 4, 5 }; + return next[st]; +} + +static inline State St_set_match(const State st) +{ + return ((st < 7) ? 7 : 10); +} + +static inline State St_set_rep(const State st) +{ + return ((st < 7) ? 8 : 11); +} + +static inline State St_set_short_rep(const State st) +{ + return ((st < 7) ? 9 : 11); +} + + +enum { + min_dictionary_bits = 12, + min_dictionary_size = 1 << min_dictionary_bits, + max_dictionary_bits = 29, + max_dictionary_size = 1 << max_dictionary_bits, + literal_context_bits = 3, + pos_state_bits = 2, + pos_states = 1 << pos_state_bits, + pos_state_mask = pos_states - 1, + + len_states = 4, + dis_slot_bits = 6, + start_dis_model = 4, + end_dis_model = 14, + modeled_distances = 1 << (end_dis_model / 2), /* 128 */ + dis_align_bits = 4, + dis_align_size = 1 << dis_align_bits, + + len_low_bits = 3, + len_mid_bits = 3, + len_high_bits = 8, + len_low_symbols = 1 << len_low_bits, + len_mid_symbols = 1 << len_mid_bits, + len_high_symbols = 1 << len_high_bits, + max_len_symbols = len_low_symbols + len_mid_symbols + len_high_symbols, + + min_match_len = 2, /* must be 2 */ + max_match_len = min_match_len + max_len_symbols - 1, /* 273 */ + min_match_len_limit = 5 +}; + +static inline int get_len_state(const int len) +{ + return min(len - min_match_len, len_states - 1); +} + +static inline int get_lit_state(const uint8_t prev_byte) +{ + return (prev_byte >> (8 - literal_context_bits)); +} + + +enum { bit_model_move_bits = 5, + bit_model_total_bits = 11, + bit_model_total = 1 << bit_model_total_bits +}; + +typedef int Bit_model; + +static inline void Bm_init(Bit_model * const probability) +{ + *probability = bit_model_total / 2; +} + +static inline void Bm_array_init(Bit_model bm[], const int size) +{ + int i; + + for (i = 0; i < size; ++i) + Bm_init(&bm[i]); +} + +struct Len_model { + Bit_model choice1; + Bit_model choice2; + Bit_model bm_low[pos_states][len_low_symbols]; + Bit_model bm_mid[pos_states][len_mid_symbols]; + Bit_model bm_high[len_high_symbols]; +}; + +static inline void Lm_init(struct Len_model * const lm) +{ + Bm_init(&lm->choice1); + Bm_init(&lm->choice2); + Bm_array_init(lm->bm_low[0], pos_states * len_low_symbols); + Bm_array_init(lm->bm_mid[0], pos_states * len_mid_symbols); + Bm_array_init(lm->bm_high, len_high_symbols); +} + + +/* Table of CRCs of all 8-bit messages. */ +STATIC_RW_DATA const uint32_t crc32[256] = + { + 0x00000000, 0x77073096, 0xEE0E612C, 0x990951BA, 0x076DC419, 0x706AF48F, + 0xE963A535, 0x9E6495A3, 0x0EDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988, + 0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91, 0x1DB71064, 0x6AB020F2, + 0xF3B97148, 0x84BE41DE, 0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7, + 0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC, 0x14015C4F, 0x63066CD9, + 0xFA0F3D63, 0x8D080DF5, 0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172, + 0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B, 0x35B5A8FA, 0x42B2986C, + 0xDBBBC9D6, 0xACBCF940, 0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59, + 0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116, 0x21B4F4B5, 0x56B3C423, + 0xCFBA9599, 0xB8BDA50F, 0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924, + 0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D, 0x76DC4190, 0x01DB7106, + 0x98D220BC, 0xEFD5102A, 0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433, + 0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818, 0x7F6A0DBB, 0x086D3D2D, + 0x91646C97, 0xE6635C01, 0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E, + 0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457, 0x65B0D9C6, 0x12B7E950, + 0x8BBEB8EA, 0xFCB9887C, 0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65, + 0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2, 0x4ADFA541, 0x3DD895D7, + 0xA4D1C46D, 0xD3D6F4FB, 0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0, + 0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7CC9, 0x5005713C, 0x270241AA, + 0xBE0B1010, 0xC90C2086, 0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F, + 0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4, 0x59B33D17, 0x2EB40D81, + 0xB7BD5C3B, 0xC0BA6CAD, 0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A, + 0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683, 0xE3630B12, 0x94643B84, + 0x0D6D6A3E, 0x7A6A5AA8, 0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1, + 0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE, 0xF762575D, 0x806567CB, + 0x196C3671, 0x6E6B06E7, 0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC, + 0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5, 0xD6D6A3E8, 0xA1D1937E, + 0x38D8C2C4, 0x4FDFF252, 0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B, + 0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60, 0xDF60EFC3, 0xA867DF55, + 0x316E8EEF, 0x4669BE79, 0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236, + 0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F, 0xC5BA3BBE, 0xB2BD0B28, + 0x2BB45A92, 0x5CB36A04, 0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D, + 0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A, 0x9C0906A9, 0xEB0E363F, + 0x72076785, 0x05005713, 0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38, + 0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21, 0x86D3D2D4, 0xF1D4E242, + 0x68DDB3F8, 0x1FDA836E, 0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777, + 0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C, 0x8F659EFF, 0xF862AE69, + 0x616BFFD3, 0x166CCF45, 0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2, + 0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB, 0xAED16A4A, 0xD9D65ADC, + 0x40DF0B66, 0x37D83BF0, 0xA9BCAE53, 0xDEBB9EC5, 0x47B2CF7F, 0x30B5FFE9, + 0xBDBDF21C, 0xCABAC28A, 0x53B39330, 0x24B4A3A6, 0xBAD03605, 0xCDD70693, + 0x54DE5729, 0x23D967BF, 0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94, + 0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D }; + + +static inline void CRC32_update_buf(uint32_t * const crc, + const uint8_t * const buffer, + const long size) +{ + long i; + uint32_t c = *crc; + + for (i = 0; i < size; ++i) + c = crc32[(c^buffer[i])&0xFF] ^ (c >> 8); + *crc = c; +} + + +STATIC_RW_DATA const uint8_t lzip_magic[4] = { 0x4C, 0x5A, 0x49, 0x50 }; /* "LZIP" */ + +typedef uint8_t Lzip_header[6]; /* 0-3 magic bytes */ + /* 4 version */ + /* 5 coded_dict_size */ +enum { Lh_size = 6 }; + +static inline bool Lh_verify_magic(const Lzip_header data) +{ + int i; + + for (i = 0; i < 4; ++i) + if (data[i] != lzip_magic[i]) + return false; + return true; +} + +/* detect (truncated) header */ +static inline bool Lh_verify_prefix(const Lzip_header data, const int sz) +{ + int i; + for (i = 0; i < sz && i < 4; ++i) + if (data[i] != lzip_magic[i]) + return false; + return (sz > 0); +} + +/* detect corrupt header */ +static inline bool Lh_verify_corrupt(const Lzip_header data) +{ + int matches = 0; + int i; + for (i = 0; i < 4; ++i) + if (data[i] == lzip_magic[i]) + ++matches; + return (matches > 1 && matches < 4); +} + +static inline bool Lh_verify_version(const Lzip_header data) +{ + return (data[4] == 1); +} + +static inline unsigned Lh_get_dictionary_size(const Lzip_header data) +{ + unsigned sz = (1 << (data[5] & 0x1F)); + + if (sz > min_dictionary_size) + sz -= (sz / 16) * ((data[5] >> 5) & 7); + return sz; +} + + +typedef uint8_t Lzip_trailer[20]; + /* 0-3 CRC32 of the uncompressed data */ + /* 4-11 size of the uncompressed data */ + /* 12-19 member size including header and trailer */ +enum { Lt_size = 20 }; + +static inline unsigned Lt_get_data_crc(const Lzip_trailer data) +{ + unsigned tmp = 0; + int i; + + for (i = 3; i >= 0; --i) { + tmp <<= 8; + tmp += data[i]; + } + return tmp; +} + +static inline unsigned long long Lt_get_data_size(const Lzip_trailer data) +{ + unsigned long long tmp = 0; + int i; + + for (i = 11; i >= 4; --i) { + tmp <<= 8; + tmp += data[i]; + } + return tmp; +} + +static inline unsigned long long Lt_get_member_size(const Lzip_trailer data) +{ + unsigned long long tmp = 0; + int i; + + for (i = 19; i >= 12; --i) { + tmp <<= 8; + tmp += data[i]; + } + return tmp; +} + + +struct Range_decoder { + unsigned long long partial_member_pos; + uint8_t *buffer; /* input buffer */ + long buffer_size; + long pos; /* current pos in buffer */ + long stream_pos; /* when reached, a new block must be read */ + uint32_t code; + uint32_t range; + long (*fill)(void*, unsigned long); + bool at_stream_end; + bool buffer_given; +}; + + +static bool Rd_read_block(struct Range_decoder * const rdec) +{ + if (!rdec->at_stream_end) { + rdec->stream_pos = rdec->fill ? + rdec->fill(rdec->buffer, rdec->buffer_size) : 0; + rdec->at_stream_end = (rdec->stream_pos < rdec->buffer_size); + rdec->partial_member_pos += rdec->pos; + rdec->pos = 0; + } + return rdec->pos < rdec->stream_pos; +} + + +static inline bool Rd_init(struct Range_decoder * const rdec, + uint8_t * const inbuf, const long in_len, + long (*fill)(void*, unsigned long)) +{ + rdec->partial_member_pos = 0; + rdec->buffer_given = (inbuf && in_len > 0); + rdec->buffer_size = rdec->buffer_given ? in_len : 16384; + rdec->buffer = rdec->buffer_given ? inbuf : malloc(rdec->buffer_size); + if (!rdec->buffer) + return false; + rdec->pos = 0; + rdec->stream_pos = rdec->buffer_given ? in_len : 0; + rdec->code = 0; + rdec->range = 0xFFFFFFFFU; + rdec->fill = fill; + rdec->at_stream_end = false; + return true; +} + +static inline void Rd_free(struct Range_decoder * const rdec) +{ + if (!rdec->buffer_given) + free(rdec->buffer); +} + +static inline bool Rd_finished(struct Range_decoder * const rdec) +{ + return rdec->pos >= rdec->stream_pos && !Rd_read_block(rdec); +} + +static inline unsigned long long +Rd_member_position(const struct Range_decoder * const rdec) +{ + return rdec->partial_member_pos + rdec->pos; +} + +static inline void Rd_reset_member_position(struct Range_decoder * const rdec) +{ + rdec->partial_member_pos = 0; rdec->partial_member_pos -= rdec->pos; +} + +static inline uint8_t Rd_get_byte(struct Range_decoder * const rdec) +{ + /* 0xFF avoids decoder error if member is truncated at EOS marker */ + if (Rd_finished(rdec)) + return 0xFF; + return rdec->buffer[rdec->pos++]; +} + +static inline void Rd_load(struct Range_decoder * const rdec) +{ + int i; + + rdec->code = 0; + for (i = 0; i < 5; ++i) + rdec->code = (rdec->code << 8) | Rd_get_byte(rdec); + rdec->range = 0xFFFFFFFFU; +} + +static inline void Rd_normalize(struct Range_decoder * const rdec) +{ + if (rdec->range <= 0x00FFFFFFU) { + rdec->range <<= 8; + rdec->code = (rdec->code << 8) | Rd_get_byte(rdec); + } +} + +static inline unsigned Rd_decode(struct Range_decoder * const rdec, + const int num_bits) +{ + unsigned symbol = 0; + int i; + + for (i = num_bits; i > 0; --i) { + bool bit; + + Rd_normalize(rdec); + rdec->range >>= 1; + /* symbol <<= 1; */ + /* if(rdec->code >= rdec->range) { rdec->code -= rdec->range; symbol |= 1; } */ + bit = (rdec->code >= rdec->range); + symbol = (symbol << 1) + bit; + rdec->code -= rdec->range & (0U - bit); + } + return symbol; +} + +static inline unsigned Rd_decode_bit(struct Range_decoder * const rdec, + Bit_model * const probability) +{ + uint32_t bound; + + Rd_normalize(rdec); + bound = (rdec->range >> bit_model_total_bits) * *probability; + if (rdec->code < bound) { + rdec->range = bound; + *probability += (bit_model_total - *probability) >> bit_model_move_bits; + return 0; + } else { + rdec->range -= bound; + rdec->code -= bound; + *probability -= *probability >> bit_model_move_bits; + return 1; + } +} + +static inline unsigned Rd_decode_tree3(struct Range_decoder * const rdec, + Bit_model bm[]) +{ + unsigned symbol = 1; + + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + return symbol & 7; +} + +static inline unsigned Rd_decode_tree6(struct Range_decoder * const rdec, + Bit_model bm[]) +{ + unsigned symbol = 1; + + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + return symbol & 0x3F; +} + +static inline unsigned Rd_decode_tree8(struct Range_decoder * const rdec, + Bit_model bm[]) +{ + unsigned symbol = 1; + int i; + + for (i = 0; i < 8; ++i) + symbol = (symbol << 1) | Rd_decode_bit(rdec, &bm[symbol]); + return symbol & 0xFF; +} + +static inline unsigned +Rd_decode_tree_reversed(struct Range_decoder * const rdec, + Bit_model bm[], const int num_bits) +{ + unsigned model = 1; + unsigned symbol = 0; + int i; + + for (i = 0; i < num_bits; ++i) { + const unsigned bit = Rd_decode_bit(rdec, &bm[model]); + + model = (model << 1) + bit; + symbol |= (bit << i); + } + return symbol; +} + +static inline unsigned +Rd_decode_tree_reversed4(struct Range_decoder * const rdec, Bit_model bm[]) +{ + unsigned symbol = Rd_decode_bit(rdec, &bm[1]); + unsigned model = 2 + symbol; + unsigned bit = Rd_decode_bit(rdec, &bm[model]); + + model = (model << 1) + bit; symbol |= (bit << 1); + bit = Rd_decode_bit(rdec, &bm[model]); + model = (model << 1) + bit; symbol |= (bit << 2); + symbol |= (Rd_decode_bit(rdec, &bm[model]) << 3); + return symbol; +} + +static inline unsigned Rd_decode_matched(struct Range_decoder * const rdec, + Bit_model bm[], unsigned match_byte) +{ + unsigned symbol = 1; + unsigned mask = 0x100; + + while (true) { + const unsigned match_bit = (match_byte <<= 1) & mask; + const unsigned bit = Rd_decode_bit(rdec, &bm[symbol+match_bit+mask]); + + symbol = (symbol << 1) + bit; + if (symbol > 0xFF) + return symbol & 0xFF; + mask &= ~(match_bit ^ (bit << 8)); /* if( match_bit != bit ) mask = 0; */ + } +} + +static inline unsigned Rd_decode_len(struct Range_decoder * const rdec, + struct Len_model * const lm, + const int pos_state) +{ + if (Rd_decode_bit(rdec, &lm->choice1) == 0) + return Rd_decode_tree3(rdec, lm->bm_low[pos_state]); + if (Rd_decode_bit(rdec, &lm->choice2) == 0) + return len_low_symbols + + Rd_decode_tree3(rdec, lm->bm_mid[pos_state]); + return len_low_symbols + len_mid_symbols + + Rd_decode_tree8(rdec, lm->bm_high); +} + + +struct LZ_decoder { + unsigned long long partial_data_pos; + struct Range_decoder *rdec; + /* Don't move bm_* to LZd_decode_member; makes frame too large. */ + Bit_model bm_literal[1 << literal_context_bits][0x300]; + Bit_model bm_match[states][pos_states]; + Bit_model bm_rep[states]; + Bit_model bm_rep0[states]; + Bit_model bm_rep1[states]; + Bit_model bm_rep2[states]; + Bit_model bm_len[states][pos_states]; + Bit_model bm_dis_slot[len_states][1 << dis_slot_bits]; + Bit_model bm_dis[modeled_distances-end_dis_model+1]; + Bit_model bm_align[dis_align_size]; + struct Len_model match_len_model; + struct Len_model rep_len_model; + + unsigned long buffer_size; + unsigned dictionary_size; + uint8_t *buffer; /* output buffer */ + unsigned long pos; /* current pos in buffer */ + unsigned long stream_pos; /* first byte not yet written to file */ + uint32_t crc; + long (*flush)(void*, unsigned long); + bool pos_wrapped; + bool buffer_given; + bool write_error; +}; + +static void LZd_flush_data(struct LZ_decoder * const d) +{ + if (d->pos > d->stream_pos) { + const long size = d->pos - d->stream_pos; + + CRC32_update_buf(&d->crc, d->buffer + d->stream_pos, size); + if ((d->flush && + d->flush(d->buffer + d->stream_pos, size) != size) || + (!d->flush && d->pos_wrapped)) + d->write_error = true; + if (d->pos >= d->buffer_size) { + d->partial_data_pos += d->pos; + d->pos = 0; + d->pos_wrapped = true; + } + d->stream_pos = d->pos; + } +} + +static inline uint8_t LZd_peek_prev(const struct LZ_decoder * const d) +{ + if (d->pos > 0) + return d->buffer[d->pos-1]; + if (d->pos_wrapped) + return d->buffer[d->buffer_size-1]; + return 0; /* prev_byte of first byte */ +} + +static inline uint8_t LZd_peek(const struct LZ_decoder * const d, + const unsigned distance) +{ + const unsigned long i = ((d->pos > distance) ? 0 : d->buffer_size) + + d->pos - distance - 1; + return d->buffer[i]; +} + +static inline void LZd_put_byte(struct LZ_decoder * const d, const uint8_t b) +{ + d->buffer[d->pos] = b; + if (++d->pos >= d->buffer_size) + LZd_flush_data(d); +} + +static inline void LZd_copy_block(struct LZ_decoder * const d, + const unsigned distance, unsigned len) +{ + unsigned long lpos = d->pos, i = lpos - distance - 1; + bool fast, fast2; + + if (lpos > distance) { + fast = (len < d->buffer_size - lpos); + fast2 = (fast && len <= lpos - i); + } else { + i += d->buffer_size; + fast = (len < d->buffer_size - i); /* (i == pos) may happen */ + fast2 = (fast && len <= i - lpos); + } + if (fast) { /* no wrap */ + d->pos += len; + if (fast2) /* no wrap, no overlap */ + memcpy(d->buffer + lpos, d->buffer + i, len); + else + for (; len > 0; --len) + d->buffer[lpos++] = d->buffer[i++]; + } else + for (; len > 0; --len) { + d->buffer[d->pos] = d->buffer[i]; + if (++d->pos >= d->buffer_size) + LZd_flush_data(d); + if (++i >= d->buffer_size) + i = 0; + } +} + +static inline bool LZd_init(struct LZ_decoder * const d, + struct Range_decoder * const rde, + const unsigned dict_size, uint8_t * const outbuf, + long out_size, long (*flush)(void*, unsigned long)) +{ + d->partial_data_pos = 0; + d->rdec = rde; + Bm_array_init(d->bm_literal[0], (1 << literal_context_bits) * 0x300); + Bm_array_init(d->bm_match[0], states * pos_states); + Bm_array_init(d->bm_rep, states); + Bm_array_init(d->bm_rep0, states); + Bm_array_init(d->bm_rep1, states); + Bm_array_init(d->bm_rep2, states); + Bm_array_init(d->bm_len[0], states * pos_states); + Bm_array_init(d->bm_dis_slot[0], len_states * (1 << dis_slot_bits)); + Bm_array_init(d->bm_dis, modeled_distances - end_dis_model + 1); + Bm_array_init(d->bm_align, dis_align_size); + Lm_init(&d->match_len_model); + Lm_init(&d->rep_len_model); + + d->buffer_given = (outbuf && out_size > 0); + d->buffer_size = d->buffer_given ? out_size : dict_size; + d->dictionary_size = min_t(unsigned long, d->buffer_size, dict_size); + d->buffer = d->buffer_given ? outbuf : large_malloc(d->buffer_size); + if (!d->buffer) + return false; + d->pos = 0; + d->stream_pos = 0; + d->crc = 0xFFFFFFFFU; + d->flush = flush; + d->pos_wrapped = false; + d->write_error = false; + /* prev_byte of first byte; also for LZd_peek( 0 ) on corrupt file */ + if (!d->buffer_given) /* inbuf and outbuf may overlap */ + d->buffer[d->buffer_size-1] = 0; + return true; +} + +static inline void LZd_free(struct LZ_decoder * const d) +{ + if (!d->buffer_given) + large_free(d->buffer); +} + +static inline unsigned LZd_crc(const struct LZ_decoder * const d) +{ + return d->crc ^ 0xFFFFFFFFU; +} + +static inline unsigned long long +LZd_data_position(const struct LZ_decoder * const d) +{ + return d->partial_data_pos + d->pos; +} + + +static bool LZd_verify_trailer(struct LZ_decoder * const d) +{ + Lzip_trailer trailer; + int i = 0; + + while (i < Lt_size) + trailer[i++] = Rd_get_byte(d->rdec); + + return (Lt_get_data_crc(trailer) == LZd_crc(d) && + Lt_get_data_size(trailer) == LZd_data_position(d) && + Lt_get_member_size(trailer) == Rd_member_position(d->rdec)); +} + + +/* Return value: 0 = OK, < 0 = error (see <linux/lzip.h>). */ +static int LZd_decode_member(struct LZ_decoder * const d) +{ + struct Range_decoder * const rdec = d->rdec; + unsigned rep0 = 0; /* rep[0-3] latest four distances */ + unsigned rep1 = 0; /* used for efficient coding of */ + unsigned rep2 = 0; /* repeated distances */ + unsigned rep3 = 0; + State state = 0; + + Rd_load(rdec); + while (!Rd_finished(rdec)) { + int len; + const int pos_state = LZd_data_position(d) & pos_state_mask; + + if (Rd_decode_bit(rdec, &d->bm_match[state][pos_state]) == 0) { + /* literal byte */ + Bit_model * const bm = d->bm_literal[get_lit_state(LZd_peek_prev(d))]; + + if (St_is_char(state)) { + state -= (state < 4) ? state : 3; + LZd_put_byte(d, Rd_decode_tree8(rdec, bm)); + } else { + state -= (state < 10) ? 3 : 6; + LZd_put_byte(d, Rd_decode_matched(rdec, bm, LZd_peek(d, rep0))); + } + continue; + } + /* match or repeated match */ + if (Rd_decode_bit(rdec, &d->bm_rep[state]) != 0) { + if (Rd_decode_bit(rdec, &d->bm_rep0[state]) == 0) { + if (Rd_decode_bit(rdec, &d->bm_len[state][pos_state]) == 0) { + state = St_set_short_rep(state); + LZd_put_byte(d, LZd_peek(d, rep0)); + continue; + } + } else { + unsigned distance; + + if (Rd_decode_bit(rdec, &d->bm_rep1[state]) == 0) + distance = rep1; + else { + if (Rd_decode_bit(rdec, &d->bm_rep2[state]) == 0) + distance = rep2; + else { + distance = rep3; + rep3 = rep2; + } + rep2 = rep1; + } + rep1 = rep0; + rep0 = distance; + } + state = St_set_rep(state); + len = min_match_len + Rd_decode_len(rdec, &d->rep_len_model, pos_state); + } else { /* match */ + unsigned distance; + + len = min_match_len + Rd_decode_len(rdec, &d->match_len_model, pos_state); + distance = Rd_decode_tree6(rdec, d->bm_dis_slot[get_len_state(len)]); + if (distance >= start_dis_model) { + const unsigned dis_slot = distance; + const int direct_bits = (dis_slot >> 1) - 1; + + distance = (2 | (dis_slot & 1)) << direct_bits; + if (dis_slot < end_dis_model) + distance += Rd_decode_tree_reversed(rdec, + d->bm_dis + (distance - dis_slot), direct_bits); + else { + distance += + Rd_decode(rdec, direct_bits - dis_align_bits) << dis_align_bits; + distance += Rd_decode_tree_reversed4(rdec, d->bm_align); + if (distance == 0xFFFFFFFFU) { /* marker found */ + Rd_normalize(rdec); + LZd_flush_data(d); + if (d->write_error) + return LZIP_WRITE_ERROR; + if (len == min_match_len) { /* End Of Stream marker */ + if (LZd_verify_trailer(d)) + return 0; + else + return LZIP_BAD_CRC; + } + if (len == min_match_len + 1) { /* Sync Flush marker */ + Rd_load(rdec); + continue; + } + return LZIP_BAD_DATA; /* unknown marker */ + } + } + } + rep3 = rep2; rep2 = rep1; rep1 = rep0; rep0 = distance; + state = St_set_match(state); + if (rep0 >= d->dictionary_size || + (rep0 >= d->pos && !d->pos_wrapped)) { + LZd_flush_data(d); + return LZIP_BAD_DATA; + } + } + LZd_copy_block(d, rep0, len); + } + LZd_flush_data(d); + return LZIP_DATA_EOF; +} + + +int lzip_decompress(unsigned char *inbuf, long in_len, + long (*fill)(void*, unsigned long), + long (*flush)(void*, unsigned long), + unsigned char *outbuf, long out_size, + long *in_posp, long *out_posp) +{ + unsigned char *outptr = outbuf; + struct Range_decoder rdec; + struct LZ_decoder *decoder = 0; + int retval = 0; + bool first_member; + + if (in_posp) + *in_posp = 0; + if (out_posp) + *out_posp = 0; + + if (!Rd_init(&rdec, inbuf, in_len, fill)) + return LZIP_OOM_INBUF; + + for (first_member = true;; first_member = false) { + long data_pos; + int size; + unsigned dictionary_size; + Lzip_header header; + + Rd_reset_member_position(&rdec); + for (size = 0; size < Lh_size && !Rd_finished(&rdec); ++size) + header[size] = Rd_get_byte(&rdec); + if (Rd_finished(&rdec)) { /* End Of File */ + if (first_member) + retval = LZIP_HEADER1_EOF; + else if (Lh_verify_prefix(header, size)) + retval = LZIP_HEADER2_EOF; + break; + } + if (!Lh_verify_magic(header)) { + if (first_member) + retval = LZIP_BAD_MAGIC1; + else if (Lh_verify_corrupt(header)) + retval = LZIP_BAD_MAGIC2; + break; + } + if (!Lh_verify_version(header)) { + retval = LZIP_BAD_VERSION; + break; + } + dictionary_size = Lh_get_dictionary_size(header); + if (dictionary_size < min_dictionary_size || + dictionary_size > max_dictionary_size) { + retval = LZIP_BAD_DICT_SIZE; + break; + } + + if (!decoder) + decoder = malloc(sizeof *decoder); + if (!decoder || !LZd_init(decoder, &rdec, dictionary_size, + outptr, out_size, flush)) { + retval = LZIP_OOM_OUTBUF; + break; + } + retval = LZd_decode_member(decoder); + if (in_posp) + *in_posp += Rd_member_position(&rdec); + data_pos = LZd_data_position(decoder); + if (outptr) + outptr += data_pos; + if (out_posp) + *out_posp += data_pos; + if (out_size > 0) + out_size -= data_pos; + LZd_free(decoder); + if (retval != 0) + break; + } + if (decoder) + free(decoder); + Rd_free(&rdec); + return retval; +} + +#ifndef STATIC +EXPORT_SYMBOL_GPL(lzip_decompress); +MODULE_DESCRIPTION("LZIP Decompressor"); +MODULE_AUTHOR("Antonio Diaz Diaz <antonio@gnu.org>"); +MODULE_LICENSE("GPL"); +#endif |