deflate.c

Lib files of linux kernel

/* +++ deflate.c */
/* deflate.c -- compress data using the deflation algorithm
 * Copyright (C) 1995-1996 Jean-loup Gailly.
 * For conditions of distribution and use, see copyright notice in zlib.h 
 */

/*
 *  ALGORITHM
 *
 *      The "deflation" process depends on being able to identify portions
 *      of the input text which are identical to earlier input (within a
 *      sliding window trailing behind the input currently being processed).
 *
 *      The most straightforward technique turns out to be the fastest for
 *      most input files: try all possible matches and select the longest.
 *      The key feature of this algorithm is that insertions into the string
 *      dictionary are very simple and thus fast, and deletions are avoided
 *      completely. Insertions are performed at each input character, whereas
 *      string matches are performed only when the previous match ends. So it
 *      is preferable to spend more time in matches to allow very fast string
 *      insertions and avoid deletions. The matching algorithm for small
 *      strings is inspired from that of Rabin & Karp. A brute force approach
 *      is used to find longer strings when a small match has been found.
 *      A similar algorithm is used in comic (by Jan-Mark Wams) and freeze
 *      (by Leonid Broukhis).
 *         A previous version of this file used a more sophisticated algorithm
 *      (by Fiala and Greene) which is guaranteed to run in linear amortized
 *      time, but has a larger average cost, uses more memory and is patented.
 *      However the F&G algorithm may be faster for some highly redundant
 *      files if the parameter max_chain_length (described below) is too large.
 *
 *  ACKNOWLEDGEMENTS
 *
 *      The idea of lazy evaluation of matches is due to Jan-Mark Wams, and
 *      I found it in 'freeze' written by Leonid Broukhis.
 *      Thanks to many people for bug reports and testing.
 *
 *  REFERENCES
 *
 *      Deutsch, L.P.,"DEFLATE Compressed Data Format Specification".
 *      Available in ftp://ds.internic.net/rfc/rfc1951.txt
 *
 *      A description of the Rabin and Karp algorithm is given in the book
 *         "Algorithms" by R. Sedgewick, Addison-Wesley, p252.
 *
 *      Fiala,E.R., and Greene,D.H.
 *         Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595
 *
 */

#include <linux/module.h>
#include <linux/zutil.h>
#include "defutil.h"


/* ===========================================================================
 *  Function prototypes.
 */
typedef enum {
    need_more,      /* block not completed, need more input or more output */
    block_done,     /* block flush performed */
    finish_started, /* finish started, need only more output at next deflate */
    finish_done     /* finish done, accept no more input or output */
} block_state;

typedef block_state (*compress_func) (deflate_state *s, int flush);
/* Compression function. Returns the block state after the call. */

static void fill_window    (deflate_state *s);
static block_state deflate_stored (deflate_state *s, int flush);
static block_state deflate_fast   (deflate_state *s, int flush);
static block_state deflate_slow   (deflate_state *s, int flush);
static void lm_init        (deflate_state *s);
static void putShortMSB    (deflate_state *s, uInt b);
static void flush_pending  (z_streamp strm);
static int read_buf        (z_streamp strm, Byte *buf, unsigned size);
static uInt longest_match  (deflate_state *s, IPos cur_match);

#ifdef DEBUG_ZLIB
static  void check_match (deflate_state *s, IPos start, IPos match,
                         int length);
#endif

/* ===========================================================================
 * Local data
 */

#define NIL 0
/* Tail of hash chains */

#ifndef TOO_FAR
#  define TOO_FAR 4096
#endif
/* Matches of length 3 are discarded if their distance exceeds TOO_FAR */

#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1)
/* Minimum amount of lookahead, except at the end of the input file.
 * See deflate.c for comments about the MIN_MATCH+1.
 */

/* Values for max_lazy_match, good_match and max_chain_length, depending on
 * the desired pack level (0..9). The values given below have been tuned to
 * exclude worst case performance for pathological files. Better values may be
 * found for specific files.
 */
typedef struct config_s {
   ush good_length; /* reduce lazy search above this match length */
   ush max_lazy;    /* do not perform lazy search above this match length */
   ush nice_length; /* quit search above this match length */
   ush max_chain;
   compress_func func;
} config;

static const config configuration_table[10] = {
/*      good lazy nice chain */
/* 0 */ {0,    0,  0,    0, deflate_stored},  /* store only */
/* 1 */ {4,    4,  8,    4, deflate_fast}, /* maximum speed, no lazy matches */
/* 2 */ {4,    5, 16,    8, deflate_fast},
/* 3 */ {4,    6, 32,   32, deflate_fast},

/* 4 */ {4,    4, 16,   16, deflate_slow},  /* lazy matches */
/* 5 */ {8,   16, 32,   32, deflate_slow},
/* 6 */ {8,   16, 128, 128, deflate_slow},
/* 7 */ {8,   32, 128, 256, deflate_slow},
/* 8 */ {32, 128, 258, 1024, deflate_slow},
/* 9 */ {32, 258, 258, 4096, deflate_slow}}; /* maximum compression */

/* Note: the deflate() code requires max_lazy >= MIN_MATCH and max_chain >= 4
 * For deflate_fast() (levels <= 3) good is ignored and lazy has a different
 * meaning.
 */

#define EQUAL 0
/* result of memcmp for equal strings */

/* ===========================================================================
 * Update a hash value with the given input byte
 * IN  assertion: all calls to to UPDATE_HASH are made with consecutive
 *    input characters, so that a running hash key can be computed from the
 *    previous key instead of complete recalculation each time.
 */
#define UPDATE_HASH(s,h,c) (h = (((h)<<s->hash_shift) ^ (c)) & s->hash_mask)


/* ===========================================================================
 * Insert string str in the dictionary and set match_head to the previous head
 * of the hash chain (the most recent string with same hash key). Return
 * the previous length of the hash chain.
 * IN  assertion: all calls to to INSERT_STRING are made with consecutive
 *    input characters and the first MIN_MATCH bytes of str are valid
 *    (except for the last MIN_MATCH-1 bytes of the input file).
 */
#define INSERT_STRING(s, str, match_head) \
   (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \
    s->prev[(str) & s->w_mask] = match_head = s->head[s->ins_h], \
    s->head[s->ins_h] = (Pos)(str))

/* ===========================================================================
 * Initialize the hash table (avoiding 64K overflow for 16 bit systems).
 * prev[] will be initialized on the fly.
 */
#define CLEAR_HASH(s) \
    s->head[s->hash_size-1] = NIL; \
    memset((char *)s->head, 0, (unsigned)(s->hash_size-1)*sizeof(*s->head));

/* ========================================================================= */
int zlib_deflateInit2(
	z_streamp strm,
	int  level,
	int  method,
	int  windowBits,
	int  memLevel,
	int  strategy
)
{
    deflate_state *s;
    int noheader = 0;
    deflate_workspace *mem;

    ush *overlay;
    /* We overlay pending_buf and d_buf+l_buf. This works since the average
     * output size for (length,distance) codes is <= 24 bits.
     */

    if (strm == NULL) return Z_STREAM_ERROR;

    strm->msg = NULL;

    if (level == Z_DEFAULT_COMPRESSION) level = 6;

    mem = (deflate_workspace *) strm->workspace;

    if (windowBits < 0) { /* undocumented feature: suppress zlib header */
        noheader = 1;
        windowBits = -windowBits;
    }
    if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED ||
        windowBits < 9 || windowBits > 15 || level < 0 || level > 9 ||
	strategy < 0 || strategy > Z_HUFFMAN_ONLY) {
        return Z_STREAM_ERROR;
    }
    s = (deflate_state *) &(mem->deflate_memory);
    strm->state = (struct internal_state *)s;
    s->strm = strm;

    s->noheader = noheader;
    s->w_bits = windowBits;
    s->w_size = 1 << s->w_bits;
    s->w_mask = s->w_size - 1;

    s->hash_bits = memLevel + 7;
    s->hash_size = 1 << s->hash_bits;
    s->hash_mask = s->hash_size - 1;
    s->hash_shift =  ((s->hash_bits+MIN_MATCH-1)/MIN_MATCH);

    s->window = (Byte *) mem->window_memory;
    s->prev   = (Pos *)  mem->prev_memory;
    s->head   = (Pos *)  mem->head_memory;

    s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */

    overlay = (ush *) mem->overlay_memory;
    s->pending_buf = (uch *) overlay;
    s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L);

    s->d_buf = overlay + s->lit_bufsize/sizeof(ush);
    s->l_buf = s->pending_buf + (1+sizeof(ush))*s->lit_bufsize;

    s->level = level;
    s->strategy = strategy;
    s->method = (Byte)method;

    return zlib_deflateReset(strm);
}

/* ========================================================================= */
#if 0
int zlib_deflateSetDictionary(
	z_streamp strm,
	const Byte *dictionary,
	uInt  dictLength
)
{
    deflate_state *s;
    uInt length = dictLength;
    uInt n;
    IPos hash_head = 0;

    if (strm == NULL || strm->state == NULL || dictionary == NULL)
	return Z_STREAM_ERROR;

    s = (deflate_state *) strm->state;
    if (s->status != INIT_STATE) return Z_STREAM_ERROR;

    strm->adler = zlib_adler32(strm->adler, dictionary, dictLength);

    if (length < MIN_MATCH) return Z_OK;
    if (length > MAX_DIST(s)) {
	length = MAX_DIST(s);
#ifndef USE_DICT_HEAD
	dictionary += dictLength - length; /* use the tail of the dictionary */
#endif
    }
    memcpy((char *)s->window, dictionary, length);
    s->strstart = length;
    s->block_start = (long)length;

    /* Insert all strings in the hash table (except for the last two bytes).
     * s->lookahead stays null, so s->ins_h will be recomputed at the next
     * call of fill_window.
     */
    s->ins_h = s->window[0];
    UPDATE_HASH(s, s->ins_h, s->window[1]);
    for (n = 0; n <= length - MIN_MATCH; n++) {
	INSERT_STRING(s, n, hash_head);
    }
    if (hash_head) hash_head = 0;  /* to make compiler happy */
    return Z_OK;
}
#endif  /*  0  */

/* ========================================================================= */
int zlib_deflateReset(
	z_streamp strm
)
{
    deflate_state *s;
    
    if (strm == NULL || strm->state == NULL)
        return Z_STREAM_ERROR;

    strm->total_in = strm->total_out = 0;
    strm->msg = NULL;
    strm->data_type = Z_UNKNOWN;

    s = (deflate_state *)strm->state;
    s->pending = 0;
    s->pending_out = s->pending_buf;

    if (s->noheader < 0) {
        s->noheader = 0; /* was set to -1 by deflate(..., Z_FINISH); */
    }
    s->status = s->noheader ? BUSY_STATE : INIT_STATE;
    strm->adler = 1;
    s->last_flush = Z_NO_FLUSH;

    zlib_tr_init(s);
    lm_init(s);

    return Z_OK;
}

/* ========================================================================= */
#if 0
int zlib_deflateParams(
	z_streamp strm,
	int level,
	int strategy
)
{
    deflate_state *s;
    compress_func func;
    int err = Z_OK;

    if (strm == NULL || strm->state == NULL) return Z_STREAM_ERROR;
    s = (deflate_state *) strm->state;

    if (level == Z_DEFAULT_COMPRESSION) {
	level = 6;
    }
    if (level < 0 || level > 9 || strategy < 0 || strategy > Z_HUFFMAN_ONLY) {
	return Z_STREAM_ERROR;
    }
    func = configuration_table[s->level].func;

    if (func != configuration_table[level].func && strm->total_in != 0) {
	/* Flush the last buffer: */
	err = zlib_deflate(strm, Z_PARTIAL_FLUSH);
    }
    if (s->level != level) {
	s->level = level;
	s->max_lazy_match   = configuration_table[level].max_lazy;
	s->good_match       = configuration_table[level].good_length;
	s->nice_match       = configuration_table[level].nice_length;
	s->max_chain_length = configuration_table[level].max_chain;
    }
    s->strategy = strategy;
    return err;
}
#endif  /*  0  */

/* =========================================================================
 * Put a short in the pending buffer. The 16-bit value is put in MSB order.
 * IN assertion: the stream state is correct and there is enough room in
 * pending_buf.
 */
static void putShortMSB(
	deflate_state *s,
	uInt b
)
{
    put_byte(s, (Byte)(b >> 8));
    put_byte(s, (Byte)(b & 0xff));
}   

/* =========================================================================
 * Flush as much pending output as possible. All deflate() output goes
 * through this function so some applications may wish to modify it
 * to avoid allocating a large strm->next_out buffer and copying into it.
 * (See also read_buf()).
 */
static void flush_pending(
	z_streamp strm
)
{
    deflate_state *s = (deflate_state *) strm->state;
    unsigned len = s->pending;

    if (len > strm->avail_out) len = strm->avail_out;
    if (len == 0) return;

    if (strm->next_out != NULL) {
	memcpy(strm->next_out, s->pending_out, len);
	strm->next_out += len;
    }
    s->pending_out += len;
    strm->total_out += len;
    strm->avail_out  -= len;
    s->pending -= len;
    if (s->pending == 0) {
        s->pending_out = s->pending_buf;
    }
}

/* ========================================================================= */
int zlib_deflate(
	z_streamp strm,
	int flush
)
{
    int old_flush; /* value of flush param for previous deflate call */
    deflate_state *s;

    if (strm == NULL || strm->state == NULL ||
	flush > Z_FINISH || flush < 0) {
        return Z_STREAM_ERROR;
    }
    s = (deflate_state *) strm->state;

    if ((strm->next_in == NULL && strm->avail_in != 0) ||
	(s->status == FINISH_STATE && flush != Z_FINISH)) {
        return Z_STREAM_ERROR;
    }
    if (strm->avail_out == 0) return Z_BUF_ERROR;

    s->strm = strm; /* just in case */
    old_flush = s->last_flush;
    s->last_flush = flush;