Update zlib to 1.2.11

third_party/zlib/deflate.c

 /* deflate.c -- compress data using the deflation algorithm
- * Copyright (C) 1995-2013 Jean-loup Gailly and Mark Adler
+ * Copyright (C) 1995-2017 Jean-loup Gailly and Mark Adler
  * 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).
  *
@@ skipping 33 matching lines @@
  *         Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595
  *
  */
 
 /* @(#) $Id$ */
 #include <assert.h>
 #include "deflate.h"
 #include "x86.h"
 
 const char deflate_copyright[] =
-   " deflate 1.2.8 Copyright 1995-2013 Jean-loup Gailly and Mark Adler ";
+   " deflate 1.2.11 Copyright 1995-2017 Jean-loup Gailly and Mark Adler ";
 /*
   If you use the zlib library in a product, an acknowledgment is welcome
   in the documentation of your product. If for some reason you cannot
   include such an acknowledgment, I would appreciate that you keep this
   copyright string in the executable of your product.
  */
 
 /* ===========================================================================
  *  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) OF((deflate_state *s, int flush));
 /* Compression function. Returns the block state after the call. */
 
+local int deflateStateCheck      OF((z_streamp strm));
+local void slide_hash     OF((deflate_state *s));
 local void fill_window    OF((deflate_state *s));
 local block_state deflate_stored OF((deflate_state *s, int flush));
 local block_state deflate_fast   OF((deflate_state *s, int flush));
 #ifndef FASTEST
 local block_state deflate_slow   OF((deflate_state *s, int flush));
 #endif
 local block_state deflate_rle    OF((deflate_state *s, int flush));
 local block_state deflate_huff   OF((deflate_state *s, int flush));
 local void lm_init        OF((deflate_state *s));
 local void putShortMSB    OF((deflate_state *s, uInt b));
 local void flush_pending  OF((z_streamp strm));
-
+unsigned ZLIB_INTERNAL read_buf OF((z_streamp strm, Bytef *buf, unsigned size));
 #ifdef ASMV
+#  pragma message("Assembler code may have bugs -- use at your own risk")
       void match_init OF((void)); /* asm code initialization */
       uInt longest_match  OF((deflate_state *s, IPos cur_match));
 #else
 local uInt longest_match  OF((deflate_state *s, IPos cur_match));
 #endif
 
-#ifdef DEBUG
+#ifdef ZLIB_DEBUG
 local  void check_match OF((deflate_state *s, IPos start, IPos match,
                             int length));
 #endif
 
-/* For fill_window_sse.c to use */
-ZLIB_INTERNAL int read_buf OF((z_streamp strm, Bytef *buf, unsigned size));
-
 /* From crc32.c */
 extern void ZLIB_INTERNAL crc_reset(deflate_state *const s);
 extern void ZLIB_INTERNAL crc_finalize(deflate_state *const s);
 extern void ZLIB_INTERNAL copy_with_crc(z_streamp strm, Bytef *dst, long size);
 
 #ifdef _MSC_VER
 #define INLINE __inline
 #else
 #define INLINE inline
 #endif
@@ skipping 45 matching lines @@
 /* 7 */ {8,   32, 128, 256, deflate_slow},
 /* 8 */ {32, 128, 258, 1024, deflate_slow},
 /* 9 */ {32, 258, 258, 4096, deflate_slow}}; /* max compression */
 #endif
 
 /* 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 */
-
-#ifndef NO_DUMMY_DECL
-struct static_tree_desc_s {int dummy;}; /* for buggy compilers */
-#endif
-
 /* rank Z_BLOCK between Z_NO_FLUSH and Z_PARTIAL_FLUSH */
-#define RANK(f) (((f) << 1) - ((f) > 4 ? 9 : 0))
+#define RANK(f) (((f) * 2) - ((f) > 4 ? 9 : 0))
 
 /* ===========================================================================
  * 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.
+ * IN  assertion: all calls 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.
  * If this file is compiled with -DFASTEST, the compression level is forced
  * to 1, and no hash chains are maintained.
- * 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).
+ * IN  assertion: all calls 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).
  */
 local INLINE Pos insert_string_c(deflate_state *const s, const Pos str)
 {
     Pos ret;
 
     UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]);
 #ifdef FASTEST
     ret = s->head[s->ins_h];
 #else
     ret = s->prev[str & s->w_mask] = s->head[s->ins_h];
@@ skipping 12 matching lines @@
 
 
 /* ===========================================================================
  * 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; \
     zmemzero((Bytef *)s->head, (unsigned)(s->hash_size-1)*sizeof(*s->head));
 
+/* ===========================================================================
+ * Slide the hash table when sliding the window down (could be avoided with 32
+ * bit values at the expense of memory usage). We slide even when level == 0 to
+ * keep the hash table consistent if we switch back to level > 0 later.
+ */
+local void slide_hash(s)
+    deflate_state *s;
+{
+    unsigned n, m;
+    Posf *p;
+    uInt wsize = s->w_size;
+
+    n = s->hash_size;
+    p = &s->head[n];
+    do {
+        m = *--p;
+        *p = (Pos)(m >= wsize ? m - wsize : NIL);
+    } while (--n);
+    n = wsize;
+#ifndef FASTEST
+    p = &s->prev[n];
+    do {
+        m = *--p;
+        *p = (Pos)(m >= wsize ? m - wsize : NIL);
+        /* If n is not on any hash chain, prev[n] is garbage but
+         * its value will never be used.
+         */
+    } while (--n);
+#endif
+}
+
 /* ========================================================================= */
 int ZEXPORT deflateInit_(strm, level, version, stream_size)
     z_streamp strm;
     int level;
     const char *version;
     int stream_size;
 {
     return deflateInit2_(strm, level, Z_DEFLATED, MAX_WBITS, DEF_MEM_LEVEL,
                          Z_DEFAULT_STRATEGY, version, stream_size);
     /* To do: ignore strm->next_in if we use it as window */
@@ skipping 56 matching lines @@
         windowBits = -windowBits;
     }
 #ifdef GZIP
     else if (windowBits > 15) {
         wrap = 2;       /* write gzip wrapper instead */
         windowBits -= 16;
     }
 #endif
     if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED ||
         windowBits < 8 || windowBits > 15 || level < 0 || level > 9 ||
-        strategy < 0 || strategy > Z_FIXED) {
+        strategy < 0 || strategy > Z_FIXED || (windowBits == 8 && wrap != 1)) {
         return Z_STREAM_ERROR;
     }
     if (windowBits == 8) windowBits = 9;  /* until 256-byte window bug fixed */
     s = (deflate_state *) ZALLOC(strm, 1, sizeof(deflate_state));
     if (s == Z_NULL) return Z_MEM_ERROR;
     strm->state = (struct internal_state FAR *)s;
     s->strm = strm;
+    s->status = INIT_STATE;     /* to pass state test in deflateReset() */
 
     s->wrap = wrap;
     s->gzhead = Z_NULL;
-    s->w_bits = windowBits;
+    s->w_bits = (uInt)windowBits;
     s->w_size = 1 << s->w_bits;
     s->w_mask = s->w_size - 1;
 
     if (x86_cpu_enable_simd) {
         s->hash_bits = 15;
     } else {
         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 = (Bytef *) ZALLOC(strm, s->w_size + window_padding, 2*sizeof(Byte));
+    s->window = (Bytef *) ZALLOC(strm,
+                                 s->w_size + window_padding,
+                                 2*sizeof(Byte));
     s->prev   = (Posf *)  ZALLOC(strm, s->w_size, sizeof(Pos));
     s->head   = (Posf *)  ZALLOC(strm, s->hash_size, sizeof(Pos));
 
     s->high_water = 0;      /* nothing written to s->window yet */
 
     s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */
 
     overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2);
     s->pending_buf = (uchf *) overlay;
     s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L);
 
     if (s->window == Z_NULL || s->prev == Z_NULL || s->head == Z_NULL ||
         s->pending_buf == Z_NULL) {
         s->status = FINISH_STATE;
         strm->msg = ERR_MSG(Z_MEM_ERROR);
         deflateEnd (strm);
         return Z_MEM_ERROR;
     }
     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 deflateReset(strm);
 }
 
+/* =========================================================================
+ * Check for a valid deflate stream state. Return 0 if ok, 1 if not.
+ */
+local int deflateStateCheck (strm)
+    z_streamp strm;
+{
+    deflate_state *s;
+    if (strm == Z_NULL ||
+        strm->zalloc == (alloc_func)0 || strm->zfree == (free_func)0)
+        return 1;
+    s = strm->state;
+    if (s == Z_NULL || s->strm != strm || (s->status != INIT_STATE &&
+#ifdef GZIP
+                                           s->status != GZIP_STATE &&
+#endif
+                                           s->status != EXTRA_STATE &&
+                                           s->status != NAME_STATE &&
+                                           s->status != COMMENT_STATE &&
+                                           s->status != HCRC_STATE &&
+                                           s->status != BUSY_STATE &&
+                                           s->status != FINISH_STATE))
+        return 1;
+    return 0;
+}
+
 /* ========================================================================= */
 int ZEXPORT deflateSetDictionary (strm, dictionary, dictLength)
     z_streamp strm;
     const Bytef *dictionary;
     uInt  dictLength;
 {
     deflate_state *s;
     uInt str, n;
     int wrap;
     unsigned avail;
     z_const unsigned char *next;
 
-    if (strm == Z_NULL || strm->state == Z_NULL || dictionary == Z_NULL)
+    if (deflateStateCheck(strm) || dictionary == Z_NULL)
         return Z_STREAM_ERROR;
     s = strm->state;
     wrap = s->wrap;
     if (wrap == 2 || (wrap == 1 && s->status != INIT_STATE) || s->lookahead)
         return Z_STREAM_ERROR;
 
     /* when using zlib wrappers, compute Adler-32 for provided dictionary */
     if (wrap == 1)
         strm->adler = adler32(strm->adler, dictionary, dictLength);
     s->wrap = 0;                    /* avoid computing Adler-32 in read_buf */
@@ skipping 33 matching lines @@
     s->lookahead = 0;
     s->match_length = s->prev_length = MIN_MATCH-1;
     s->match_available = 0;
     strm->next_in = next;
     strm->avail_in = avail;
     s->wrap = wrap;
     return Z_OK;
 }
 
 /* ========================================================================= */
+int ZEXPORT deflateGetDictionary (strm, dictionary, dictLength)
+    z_streamp strm;
+    Bytef *dictionary;
+    uInt  *dictLength;
+{
+    deflate_state *s;
+    uInt len;
+
+    if (deflateStateCheck(strm))
+        return Z_STREAM_ERROR;
+    s = strm->state;
+    len = s->strstart + s->lookahead;
+    if (len > s->w_size)
+        len = s->w_size;
+    if (dictionary != Z_NULL && len)
+        zmemcpy(dictionary, s->window + s->strstart + s->lookahead - len, len);
+    if (dictLength != Z_NULL)
+        *dictLength = len;
+    return Z_OK;
+}
+
+/* ========================================================================= */
 int ZEXPORT deflateResetKeep (strm)
     z_streamp strm;
 {
     deflate_state *s;
 
-    if (strm == Z_NULL || strm->state == Z_NULL ||
-        strm->zalloc == (alloc_func)0 || strm->zfree == (free_func)0) {
+    if (deflateStateCheck(strm)) {
         return Z_STREAM_ERROR;
     }
 
     strm->total_in = strm->total_out = 0;
     strm->msg = Z_NULL; /* use zfree if we ever allocate msg dynamically */
     strm->data_type = Z_UNKNOWN;
 
     s = (deflate_state *)strm->state;
     s->pending = 0;
     s->pending_out = s->pending_buf;
 
     if (s->wrap < 0) {
         s->wrap = -s->wrap; /* was made negative by deflate(..., Z_FINISH); */
     }
-    s->status = s->wrap ? INIT_STATE : BUSY_STATE;
+    s->status =
+#ifdef GZIP
+        s->wrap == 2 ? GZIP_STATE :
+#endif
+        s->wrap ? INIT_STATE : BUSY_STATE;
     strm->adler =
 #ifdef GZIP
         s->wrap == 2 ? crc32(0L, Z_NULL, 0) :
 #endif
         adler32(0L, Z_NULL, 0);
     s->last_flush = Z_NO_FLUSH;
 
     _tr_init(s);
 
     return Z_OK;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflateReset (strm)
     z_streamp strm;
 {
     int ret;
 
     ret = deflateResetKeep(strm);
     if (ret == Z_OK)
         lm_init(strm->state);
     return ret;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflateSetHeader (strm, head)
     z_streamp strm;
     gz_headerp head;
 {
-    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
-    if (strm->state->wrap != 2) return Z_STREAM_ERROR;
+    if (deflateStateCheck(strm) || strm->state->wrap != 2)
+        return Z_STREAM_ERROR;
     strm->state->gzhead = head;
     return Z_OK;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflatePending (strm, pending, bits)
     unsigned *pending;
     int *bits;
     z_streamp strm;
 {
-    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
     if (pending != Z_NULL)
         *pending = strm->state->pending;
     if (bits != Z_NULL)
         *bits = strm->state->bi_valid;
     return Z_OK;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflatePrime (strm, bits, value)
     z_streamp strm;
     int bits;
     int value;
 {
     deflate_state *s;
     int put;
 
-    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
     s = strm->state;
     if ((Bytef *)(s->d_buf) < s->pending_out + ((Buf_size + 7) >> 3))
         return Z_BUF_ERROR;
     do {
         put = Buf_size - s->bi_valid;
         if (put > bits)
             put = bits;
         s->bi_buf |= (ush)((value & ((1 << put) - 1)) << s->bi_valid);
         s->bi_valid += put;
         _tr_flush_bits(s);
         value >>= put;
         bits -= put;
     } while (bits);
     return Z_OK;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflateParams(strm, level, strategy)
     z_streamp strm;
     int level;
     int strategy;
 {
     deflate_state *s;
     compress_func func;
-    int err = Z_OK;
 
-    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
     s = strm->state;
 
 #ifdef FASTEST
     if (level != 0) level = 1;
 #else
     if (level == Z_DEFAULT_COMPRESSION) level = 6;
 #endif
     if (level < 0 || level > 9 || strategy < 0 || strategy > Z_FIXED) {
         return Z_STREAM_ERROR;
     }
     func = configuration_table[s->level].func;
 
     if ((strategy != s->strategy || func != configuration_table[level].func) &&
-        strm->total_in != 0) {
+        s->high_water) {
         /* Flush the last buffer: */
-        err = deflate(strm, Z_BLOCK);
-        if (err == Z_BUF_ERROR && s->pending == 0)
-            err = Z_OK;
+        int err = deflate(strm, Z_BLOCK);
+        if (err == Z_STREAM_ERROR)
+            return err;
+        if (strm->avail_out == 0)
+            return Z_BUF_ERROR;
     }
     if (s->level != level) {
+        if (s->level == 0 && s->matches != 0) {
+            if (s->matches == 1)
+                slide_hash(s);
+            else
+                CLEAR_HASH(s);
+            s->matches = 0;
+        }
         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;
+    return Z_OK;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflateTune(strm, good_length, max_lazy, nice_length, max_chain)
     z_streamp strm;
     int good_length;
     int max_lazy;
     int nice_length;
     int max_chain;
 {
     deflate_state *s;
 
-    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
     s = strm->state;
-    s->good_match = good_length;
-    s->max_lazy_match = max_lazy;
+    s->good_match = (uInt)good_length;
+    s->max_lazy_match = (uInt)max_lazy;
     s->nice_match = nice_length;
-    s->max_chain_length = max_chain;
+    s->max_chain_length = (uInt)max_chain;
     return Z_OK;
 }
 
 /* =========================================================================
  * For the default windowBits of 15 and memLevel of 8, this function returns
  * a close to exact, as well as small, upper bound on the compressed size.
  * They are coded as constants here for a reason--if the #define's are
  * changed, then this function needs to be changed as well.  The return
  * value for 15 and 8 only works for those exact settings.
  *
  * For any setting other than those defaults for windowBits and memLevel,
  * the value returned is a conservative worst case for the maximum expansion
  * resulting from using fixed blocks instead of stored blocks, which deflate
  * can emit on compressed data for some combinations of the parameters.
  *
  * This function could be more sophisticated to provide closer upper bounds for
  * every combination of windowBits and memLevel.  But even the conservative
  * upper bound of about 14% expansion does not seem onerous for output buffer
  * allocation.
  */
 uLong ZEXPORT deflateBound(strm, sourceLen)
     z_streamp strm;
     uLong sourceLen;
 {
     deflate_state *s;
     uLong complen, wraplen;
-    Bytef *str;
 
     /* conservative upper bound for compressed data */
     complen = sourceLen +
               ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 5;
 
     /* if can't get parameters, return conservative bound plus zlib wrapper */
-    if (strm == Z_NULL || strm->state == Z_NULL)
+    if (deflateStateCheck(strm))
         return complen + 6;
 
     /* compute wrapper length */
     s = strm->state;
     switch (s->wrap) {
     case 0:                                 /* raw deflate */
         wraplen = 0;
         break;
     case 1:                                 /* zlib wrapper */
         wraplen = 6 + (s->strstart ? 4 : 0);
         break;
+#ifdef GZIP
     case 2:                                 /* gzip wrapper */
         wraplen = 18;
         if (s->gzhead != Z_NULL) {          /* user-supplied gzip header */
+            Bytef *str;
             if (s->gzhead->extra != Z_NULL)
                 wraplen += 2 + s->gzhead->extra_len;
             str = s->gzhead->name;
             if (str != Z_NULL)
                 do {
                     wraplen++;
                 } while (*str++);
             str = s->gzhead->comment;
             if (str != Z_NULL)
                 do {
                     wraplen++;
                 } while (*str++);
             if (s->gzhead->hcrc)
                 wraplen += 2;
         }
         break;
+#endif
     default:                                /* for compiler happiness */
         wraplen = 6;
     }
 
     /* if not default parameters, return conservative bound */
     if (s->w_bits != 15 || s->hash_bits != 8 + 7)
         return complen + wraplen;
 
     /* default settings: return tight bound for that case */
     return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) +
            (sourceLen >> 25) + 13 - 6 + wraplen;
 }
 
 /* =========================================================================
  * 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.
  */
 local void putShortMSB (s, b)
     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()).
+ * Flush as much pending output as possible. All deflate() output, except for
+ * some deflate_stored() 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()).
  */
 local void flush_pending(strm)
     z_streamp strm;
 {
     unsigned len;
     deflate_state *s = strm->state;
 
     _tr_flush_bits(s);
     len = s->pending;
     if (len > strm->avail_out) len = strm->avail_out;
     if (len == 0) return;
 
     zmemcpy(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;
+    strm->avail_out -= len;
+    s->pending      -= len;
     if (s->pending == 0) {
         s->pending_out = s->pending_buf;
     }
 }
 
+/* ===========================================================================
+ * Update the header CRC with the bytes s->pending_buf[beg..s->pending - 1].
+ */
+#define HCRC_UPDATE(beg) \
+    do { \
+        if (s->gzhead->hcrc && s->pending > (beg)) \
+            strm->adler = crc32(strm->adler, s->pending_buf + (beg), \
+                                s->pending - (beg)); \
+    } while (0)
+
 /* ========================================================================= */
 int ZEXPORT deflate (strm, flush)
     z_streamp strm;
     int flush;
 {
     int old_flush; /* value of flush param for previous deflate call */
     deflate_state *s;
 
-    if (strm == Z_NULL || strm->state == Z_NULL ||
-        flush > Z_BLOCK || flush < 0) {
+    if (deflateStateCheck(strm) || flush > Z_BLOCK || flush < 0) {
         return Z_STREAM_ERROR;
     }
     s = strm->state;
 
     if (strm->next_out == Z_NULL ||
-        (strm->next_in == Z_NULL && strm->avail_in != 0) ||
+        (strm->avail_in != 0 && strm->next_in == Z_NULL) ||
         (s->status == FINISH_STATE && flush != Z_FINISH)) {
         ERR_RETURN(strm, Z_STREAM_ERROR);
     }
     if (strm->avail_out == 0) ERR_RETURN(strm, Z_BUF_ERROR);
 
-    s->strm = strm; /* just in case */
     old_flush = s->last_flush;
     s->last_flush = flush;
 
-    /* Write the header */
-    if (s->status == INIT_STATE) {
-#ifdef GZIP
-        if (s->wrap == 2) {
-            crc_reset(s);
-            put_byte(s, 31);
-            put_byte(s, 139);
-            put_byte(s, 8);
-            if (s->gzhead == Z_NULL) {
-                put_byte(s, 0);
-                put_byte(s, 0);
-                put_byte(s, 0);
-                put_byte(s, 0);
-                put_byte(s, 0);
-                put_byte(s, s->level == 9 ? 2 :
-                            (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
-                             4 : 0));
-                put_byte(s, OS_CODE);
-                s->status = BUSY_STATE;
-            }
-            else {
-                put_byte(s, (s->gzhead->text ? 1 : 0) +
-                            (s->gzhead->hcrc ? 2 : 0) +
-                            (s->gzhead->extra == Z_NULL ? 0 : 4) +
-                            (s->gzhead->name == Z_NULL ? 0 : 8) +
-                            (s->gzhead->comment == Z_NULL ? 0 : 16)
-                        );
-                put_byte(s, (Byte)(s->gzhead->time & 0xff));
-                put_byte(s, (Byte)((s->gzhead->time >> 8) & 0xff));
-                put_byte(s, (Byte)((s->gzhead->time >> 16) & 0xff));
-                put_byte(s, (Byte)((s->gzhead->time >> 24) & 0xff));
-                put_byte(s, s->level == 9 ? 2 :
-                            (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
-                             4 : 0));
-                put_byte(s, s->gzhead->os & 0xff);
-                if (s->gzhead->extra != Z_NULL) {
-                    put_byte(s, s->gzhead->extra_len & 0xff);
-                    put_byte(s, (s->gzhead->extra_len >> 8) & 0xff);
-                }
-                if (s->gzhead->hcrc)
-                    strm->adler = crc32(strm->adler, s->pending_buf,
-                                        s->pending);
-                s->gzindex = 0;
-                s->status = EXTRA_STATE;
-            }
-        }
-        else
-#endif
-        {
-            uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8;
-            uInt level_flags;
-
-            if (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2)
-                level_flags = 0;
-            else if (s->level < 6)
-                level_flags = 1;
-            else if (s->level == 6)
-                level_flags = 2;
-            else
-                level_flags = 3;
-            header |= (level_flags << 6);
-            if (s->strstart != 0) header |= PRESET_DICT;
-            header += 31 - (header % 31);
-
-            s->status = BUSY_STATE;
-            putShortMSB(s, header);
-
-            /* Save the adler32 of the preset dictionary: */
-            if (s->strstart != 0) {
-                putShortMSB(s, (uInt)(strm->adler >> 16));
-                putShortMSB(s, (uInt)(strm->adler & 0xffff));
-            }
-            strm->adler = adler32(0L, Z_NULL, 0);
-        }
-    }
-#ifdef GZIP
-    if (s->status == EXTRA_STATE) {
-        if (s->gzhead->extra != Z_NULL) {
-            uInt beg = s->pending;  /* start of bytes to update crc */
-
-            while (s->gzindex < (s->gzhead->extra_len & 0xffff)) {
-                if (s->pending == s->pending_buf_size) {
-                    if (s->gzhead->hcrc && s->pending > beg)
-                        strm->adler = crc32(strm->adler, s->pending_buf + beg,
-                                            s->pending - beg);
-                    flush_pending(strm);
-                    beg = s->pending;
-                    if (s->pending == s->pending_buf_size)
-                        break;
-                }
-                put_byte(s, s->gzhead->extra[s->gzindex]);
-                s->gzindex++;
-            }
-            if (s->gzhead->hcrc && s->pending > beg)
-                strm->adler = crc32(strm->adler, s->pending_buf + beg,
-                                    s->pending - beg);
-            if (s->gzindex == s->gzhead->extra_len) {
-                s->gzindex = 0;
-                s->status = NAME_STATE;
-            }
-        }
-        else
-            s->status = NAME_STATE;
-    }
-    if (s->status == NAME_STATE) {
-        if (s->gzhead->name != Z_NULL) {
-            uInt beg = s->pending;  /* start of bytes to update crc */
-            int val;
-
-            do {
-                if (s->pending == s->pending_buf_size) {
-                    if (s->gzhead->hcrc && s->pending > beg)
-                        strm->adler = crc32(strm->adler, s->pending_buf + beg,
-                                            s->pending - beg);
-                    flush_pending(strm);
-                    beg = s->pending;
-                    if (s->pending == s->pending_buf_size) {
-                        val = 1;
-                        break;
-                    }
-                }
-                val = s->gzhead->name[s->gzindex++];
-                put_byte(s, val);
-            } while (val != 0);
-            if (s->gzhead->hcrc && s->pending > beg)
-                strm->adler = crc32(strm->adler, s->pending_buf + beg,
-                                    s->pending - beg);
-            if (val == 0) {
-                s->gzindex = 0;
-                s->status = COMMENT_STATE;
-            }
-        }
-        else
-            s->status = COMMENT_STATE;
-    }
-    if (s->status == COMMENT_STATE) {
-        if (s->gzhead->comment != Z_NULL) {
-            uInt beg = s->pending;  /* start of bytes to update crc */
-            int val;
-
-            do {
-                if (s->pending == s->pending_buf_size) {
-                    if (s->gzhead->hcrc && s->pending > beg)
-                        strm->adler = crc32(strm->adler, s->pending_buf + beg,
-                                            s->pending - beg);
-                    flush_pending(strm);
-                    beg = s->pending;
-                    if (s->pending == s->pending_buf_size) {
-                        val = 1;
-                        break;
-                    }
-                }
-                val = s->gzhead->comment[s->gzindex++];
-                put_byte(s, val);
-            } while (val != 0);
-            if (s->gzhead->hcrc && s->pending > beg)
-                strm->adler = crc32(strm->adler, s->pending_buf + beg,
-                                    s->pending - beg);
-            if (val == 0)
-                s->status = HCRC_STATE;
-        }
-        else
-            s->status = HCRC_STATE;
-    }
-    if (s->status == HCRC_STATE) {
-        if (s->gzhead->hcrc) {
-            if (s->pending + 2 > s->pending_buf_size)
-                flush_pending(strm);
-            if (s->pending + 2 <= s->pending_buf_size) {
-                put_byte(s, (Byte)(strm->adler & 0xff));
-                put_byte(s, (Byte)((strm->adler >> 8) & 0xff));
-                strm->adler = crc32(0L, Z_NULL, 0);
-                s->status = BUSY_STATE;
-            }
-        }
-        else
-            s->status = BUSY_STATE;
-    }
-#endif
-
     /* Flush as much pending output as possible */
     if (s->pending != 0) {
         flush_pending(strm);
         if (strm->avail_out == 0) {
             /* Since avail_out is 0, deflate will be called again with
              * more output space, but possibly with both pending and
              * avail_in equal to zero. There won't be anything to do,
              * but this is not an error situation so make sure we
              * return OK instead of BUF_ERROR at next call of deflate:
              */
             s->last_flush = -1;
             return Z_OK;
         }
 
     /* Make sure there is something to do and avoid duplicate consecutive
      * flushes. For repeated and useless calls with Z_FINISH, we keep
      * returning Z_STREAM_END instead of Z_BUF_ERROR.
      */
     } else if (strm->avail_in == 0 && RANK(flush) <= RANK(old_flush) &&
                flush != Z_FINISH) {
         ERR_RETURN(strm, Z_BUF_ERROR);
     }
 
     /* User must not provide more input after the first FINISH: */
     if (s->status == FINISH_STATE && strm->avail_in != 0) {
         ERR_RETURN(strm, Z_BUF_ERROR);
     }
 
+    /* Write the header */
+    if (s->status == INIT_STATE) {
+        /* zlib header */
+        uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8;
+        uInt level_flags;
+
+        if (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2)
+            level_flags = 0;
+        else if (s->level < 6)
+            level_flags = 1;
+        else if (s->level == 6)
+            level_flags = 2;
+        else
+            level_flags = 3;
+        header |= (level_flags << 6);
+        if (s->strstart != 0) header |= PRESET_DICT;
+        header += 31 - (header % 31);
+
+        putShortMSB(s, header);
+
+        /* Save the adler32 of the preset dictionary: */
+        if (s->strstart != 0) {
+            putShortMSB(s, (uInt)(strm->adler >> 16));
+            putShortMSB(s, (uInt)(strm->adler & 0xffff));
+        }
+        strm->adler = adler32(0L, Z_NULL, 0);
+        s->status = BUSY_STATE;
+
+        /* Compression must start with an empty pending buffer */
+        flush_pending(strm);
+        if (s->pending != 0) {
+            s->last_flush = -1;
+            return Z_OK;
+        }
+    }
+#ifdef GZIP
+    if (s->status == GZIP_STATE) {
+        /* gzip header */
+        crc_reset(s);
+        put_byte(s, 31);
+        put_byte(s, 139);
+        put_byte(s, 8);
+        if (s->gzhead == Z_NULL) {
+            put_byte(s, 0);
+            put_byte(s, 0);
+            put_byte(s, 0);
+            put_byte(s, 0);
+            put_byte(s, 0);
+            put_byte(s, s->level == 9 ? 2 :
+                     (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
+                      4 : 0));
+            put_byte(s, OS_CODE);
+            s->status = BUSY_STATE;
+
+            /* Compression must start with an empty pending buffer */
+            flush_pending(strm);
+            if (s->pending != 0) {
+                s->last_flush = -1;
+                return Z_OK;
+            }
+        }
+        else {
+            put_byte(s, (s->gzhead->text ? 1 : 0) +
+                     (s->gzhead->hcrc ? 2 : 0) +
+                     (s->gzhead->extra == Z_NULL ? 0 : 4) +
+                     (s->gzhead->name == Z_NULL ? 0 : 8) +
+                     (s->gzhead->comment == Z_NULL ? 0 : 16)
+                     );
+            put_byte(s, (Byte)(s->gzhead->time & 0xff));
+            put_byte(s, (Byte)((s->gzhead->time >> 8) & 0xff));
+            put_byte(s, (Byte)((s->gzhead->time >> 16) & 0xff));
+            put_byte(s, (Byte)((s->gzhead->time >> 24) & 0xff));
+            put_byte(s, s->level == 9 ? 2 :
+                     (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
+                      4 : 0));
+            put_byte(s, s->gzhead->os & 0xff);
+            if (s->gzhead->extra != Z_NULL) {
+                put_byte(s, s->gzhead->extra_len & 0xff);
+                put_byte(s, (s->gzhead->extra_len >> 8) & 0xff);
+            }
+            if (s->gzhead->hcrc)
+                strm->adler = crc32(strm->adler, s->pending_buf,
+                                    s->pending);
+            s->gzindex = 0;
+            s->status = EXTRA_STATE;
+        }
+    }
+    if (s->status == EXTRA_STATE) {
+        if (s->gzhead->extra != Z_NULL) {
+            ulg beg = s->pending;   /* start of bytes to update crc */
+            uInt left = (s->gzhead->extra_len & 0xffff) - s->gzindex;
+            while (s->pending + left > s->pending_buf_size) {
+                uInt copy = s->pending_buf_size - s->pending;
+                zmemcpy(s->pending_buf + s->pending,
+                        s->gzhead->extra + s->gzindex, copy);
+                s->pending = s->pending_buf_size;
+                HCRC_UPDATE(beg);
+                s->gzindex += copy;
+                flush_pending(strm);
+                if (s->pending != 0) {
+                    s->last_flush = -1;
+                    return Z_OK;
+                }
+                beg = 0;
+                left -= copy;
+            }
+            zmemcpy(s->pending_buf + s->pending,
+                    s->gzhead->extra + s->gzindex, left);
+            s->pending += left;
+            HCRC_UPDATE(beg);
+            s->gzindex = 0;
+        }
+        s->status = NAME_STATE;
+    }
+    if (s->status == NAME_STATE) {
+        if (s->gzhead->name != Z_NULL) {
+            ulg beg = s->pending;   /* start of bytes to update crc */
+            int val;
+            do {
+                if (s->pending == s->pending_buf_size) {
+                    HCRC_UPDATE(beg);
+                    flush_pending(strm);
+                    if (s->pending != 0) {
+                        s->last_flush = -1;
+                        return Z_OK;
+                    }
+                    beg = 0;
+                }
+                val = s->gzhead->name[s->gzindex++];
+                put_byte(s, val);
+            } while (val != 0);
+            HCRC_UPDATE(beg);
+            s->gzindex = 0;
+        }
+        s->status = COMMENT_STATE;
+    }
+    if (s->status == COMMENT_STATE) {
+        if (s->gzhead->comment != Z_NULL) {
+            ulg beg = s->pending;   /* start of bytes to update crc */
+            int val;
+            do {
+                if (s->pending == s->pending_buf_size) {
+                    HCRC_UPDATE(beg);
+                    flush_pending(strm);
+                    if (s->pending != 0) {
+                        s->last_flush = -1;
+                        return Z_OK;
+                    }
+                    beg = 0;
+                }
+                val = s->gzhead->comment[s->gzindex++];
+                put_byte(s, val);
+            } while (val != 0);
+            HCRC_UPDATE(beg);
+        }
+        s->status = HCRC_STATE;
+    }
+    if (s->status == HCRC_STATE) {
+        if (s->gzhead->hcrc) {
+            if (s->pending + 2 > s->pending_buf_size) {
+                flush_pending(strm);
+                if (s->pending != 0) {
+                    s->last_flush = -1;
+                    return Z_OK;
+                }
+            }
+            put_byte(s, (Byte)(strm->adler & 0xff));
+            put_byte(s, (Byte)((strm->adler >> 8) & 0xff));
+            strm->adler = crc32(0L, Z_NULL, 0);
+        }
+        s->status = BUSY_STATE;
+
+        /* Compression must start with an empty pending buffer */
+        flush_pending(strm);
+        if (s->pending != 0) {
+            s->last_flush = -1;
+            return Z_OK;
+        }
+    }
+#endif
+
     /* Start a new block or continue the current one.
      */
     if (strm->avail_in != 0 || s->lookahead != 0 ||
         (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) {
         block_state bstate;
 
-        bstate = s->strategy == Z_HUFFMAN_ONLY ? deflate_huff(s, flush) :
-                    (s->strategy == Z_RLE ? deflate_rle(s, flush) :
-                        (*(configuration_table[s->level].func))(s, flush));
+        bstate = s->level == 0 ? deflate_stored(s, flush) :
+                 s->strategy == Z_HUFFMAN_ONLY ? deflate_huff(s, flush) :
+                 s->strategy == Z_RLE ? deflate_rle(s, flush) :
+                 (*(configuration_table[s->level].func))(s, flush);
 
         if (bstate == finish_started || bstate == finish_done) {
             s->status = FINISH_STATE;
         }
         if (bstate == need_more || bstate == finish_started) {
             if (strm->avail_out == 0) {
                 s->last_flush = -1; /* avoid BUF_ERROR next call, see above */
             }
             return Z_OK;
             /* If flush != Z_NO_FLUSH && avail_out == 0, the next call
@@ skipping 21 matching lines @@
                     }
                 }
             }
             flush_pending(strm);
             if (strm->avail_out == 0) {
               s->last_flush = -1; /* avoid BUF_ERROR at next call, see above */
               return Z_OK;
             }
         }
     }
-    Assert(strm->avail_out > 0, "bug2");
 
     if (flush != Z_FINISH) return Z_OK;
     if (s->wrap <= 0) return Z_STREAM_END;
 
     /* Write the trailer */
 #ifdef GZIP
     if (s->wrap == 2) {
         crc_finalize(s);
         put_byte(s, (Byte)(strm->adler & 0xff));
         put_byte(s, (Byte)((strm->adler >> 8) & 0xff));
@@ skipping 17 matching lines @@
     if (s->wrap > 0) s->wrap = -s->wrap; /* write the trailer only once! */
     return s->pending != 0 ? Z_OK : Z_STREAM_END;
 }
 
 /* ========================================================================= */
 int ZEXPORT deflateEnd (strm)
     z_streamp strm;
 {
     int status;
 
-    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
 
     status = strm->state->status;
-    if (status != INIT_STATE &&
-        status != EXTRA_STATE &&
-        status != NAME_STATE &&
-        status != COMMENT_STATE &&
-        status != HCRC_STATE &&
-        status != BUSY_STATE &&
-        status != FINISH_STATE) {
-      return Z_STREAM_ERROR;
-    }
 
     /* Deallocate in reverse order of allocations: */
     TRY_FREE(strm, strm->state->pending_buf);
     TRY_FREE(strm, strm->state->head);
     TRY_FREE(strm, strm->state->prev);
     TRY_FREE(strm, strm->state->window);
 
     ZFREE(strm, strm->state);
     strm->state = Z_NULL;
 
@@ skipping 10 matching lines @@
     z_streamp source;
 {
 #ifdef MAXSEG_64K
     return Z_STREAM_ERROR;
 #else
     deflate_state *ds;
     deflate_state *ss;
     ushf *overlay;
 
 
-    if (source == Z_NULL || dest == Z_NULL || source->state == Z_NULL) {
+    if (deflateStateCheck(source) || dest == Z_NULL) {
         return Z_STREAM_ERROR;
     }
 
     ss = source->state;
 
     zmemcpy((voidpf)dest, (voidpf)source, sizeof(z_stream));
 
     ds = (deflate_state *) ZALLOC(dest, 1, sizeof(deflate_state));
     if (ds == Z_NULL) return Z_MEM_ERROR;
     dest->state = (struct internal_state FAR *) ds;
@@ skipping 29 matching lines @@
 #endif /* MAXSEG_64K */
 }
 
 /* ===========================================================================
  * Read a new buffer from the current input stream, update the adler32
  * and total number of bytes read.  All deflate() input goes through
  * this function so some applications may wish to modify it to avoid
  * allocating a large strm->next_in buffer and copying from it.
  * (See also flush_pending()).
  */
-ZLIB_INTERNAL int read_buf(strm, buf, size)
+ZLIB_INTERNAL unsigned read_buf(strm, buf, size)
     z_streamp strm;
     Bytef *buf;
     unsigned size;
 {
     unsigned len = strm->avail_in;
 
     if (len > size) len = size;
     if (len == 0) return 0;
 
     strm->avail_in  -= len;
 
 #ifdef GZIP
     if (strm->state->wrap == 2)
         copy_with_crc(strm, buf, len);
     else 
 #endif
     {
         zmemcpy(buf, strm->next_in, len);
         if (strm->state->wrap == 1)
             strm->adler = adler32(strm->adler, buf, len);
     }
     strm->next_in  += len;
     strm->total_in += len;
 
-    return (int)len;
+    return len;
 }
 
 /* ===========================================================================
  * Initialize the "longest match" routines for a new zlib stream
  */
 local void lm_init (s)
     deflate_state *s;
 {
     s->window_size = (ulg)2L*s->w_size;
 
@@ skipping 33 matching lines @@
 #ifndef ASMV
 /* For 80x86 and 680x0, an optimized version will be provided in match.asm or
  * match.S. The code will be functionally equivalent.
  */
 local uInt longest_match(s, cur_match)
     deflate_state *s;
     IPos cur_match;                             /* current match */
 {
     unsigned chain_length = s->max_chain_length;/* max hash chain length */
     register Bytef *scan = s->window + s->strstart; /* current string */
-    register Bytef *match;                       /* matched string */
+    register Bytef *match;                      /* matched string */
     register int len;                           /* length of current match */
-    int best_len = s->prev_length;              /* best match length so far */
+    int best_len = (int)s->prev_length;         /* best match length so far */
     int nice_match = s->nice_match;             /* stop if match long enough */
     IPos limit = s->strstart > (IPos)MAX_DIST(s) ?
         s->strstart - (IPos)MAX_DIST(s) : NIL;
     /* Stop when cur_match becomes <= limit. To simplify the code,
      * we prevent matches with the string of window index 0.
      */
     Posf *prev = s->prev;
     uInt wmask = s->w_mask;
 
 #ifdef UNALIGNED_OK
@@ skipping 14 matching lines @@
      */
     Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever");
 
     /* Do not waste too much time if we already have a good match: */
     if (s->prev_length >= s->good_match) {
         chain_length >>= 2;
     }
     /* Do not look for matches beyond the end of the input. This is necessary
      * to make deflate deterministic.
      */
-    if ((uInt)nice_match > s->lookahead) nice_match = s->lookahead;
+    if ((uInt)nice_match > s->lookahead) nice_match = (int)s->lookahead;
 
     Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead");
 
     do {
         Assert(cur_match < s->strstart, "no future");
         match = s->window + cur_match;
 
         /* Skip to next match if the match length cannot increase
          * or if the match length is less than 2.  Note that the checks below
          * for insufficient lookahead only occur occasionally for performance
@@ skipping 140 matching lines @@
     len = MAX_MATCH - (int)(strend - scan);
 
     if (len < MIN_MATCH) return MIN_MATCH - 1;
 
     s->match_start = cur_match;
     return (uInt)len <= s->lookahead ? (uInt)len : s->lookahead;
 }
 
 #endif /* FASTEST */
 
-#ifdef DEBUG
+#ifdef ZLIB_DEBUG
+
+#define EQUAL 0
+/* result of memcmp for equal strings */
+
 /* ===========================================================================
  * Check that the match at match_start is indeed a match.
  */
 local void check_match(s, start, match, length)
     deflate_state *s;
     IPos start, match;
     int length;
 {
     /* check that the match is indeed a match */
     if (zmemcmp(s->window + match,
                 s->window + start, length) != EQUAL) {
         fprintf(stderr, " start %u, match %u, length %d\n",
                 start, match, length);
         do {
             fprintf(stderr, "%c%c", s->window[match++], s->window[start++]);
         } while (--length != 0);
         z_error("invalid match");
     }
     if (z_verbose > 1) {
         fprintf(stderr,"\\[%d,%d]", start-match, length);
         do { putc(s->window[start++], stderr); } while (--length != 0);
     }
 }
 #else
 #  define check_match(s, start, match, length)
-#endif /* DEBUG */
+#endif /* ZLIB_DEBUG */
 
 /* ===========================================================================
  * Fill the window when the lookahead becomes insufficient.
  * Updates strstart and lookahead.
  *
  * IN assertion: lookahead < MIN_LOOKAHEAD
  * OUT assertions: strstart <= window_size-MIN_LOOKAHEAD
  *    At least one byte has been read, or avail_in == 0; reads are
  *    performed for at least two bytes (required for the zip translate_eol
  *    option -- not supported here).
  */
 local void fill_window_c(deflate_state *s);
 
 local void fill_window(deflate_state *s)
 {
     if (x86_cpu_enable_simd) {
         fill_window_sse(s);
         return;
     }
 
     fill_window_c(s);
 }
 
 local void fill_window_c(s)
     deflate_state *s;
 {
-    register unsigned n, m;
-    register Posf *p;
+    unsigned n;
     unsigned more;    /* Amount of free space at the end of the window. */
     uInt wsize = s->w_size;
 
     Assert(s->lookahead < MIN_LOOKAHEAD, "already enough lookahead");
 
     do {
         more = (unsigned)(s->window_size -(ulg)s->lookahead -(ulg)s->strstart);
 
         /* Deal with !@#$% 64K limit: */
         if (sizeof(int) <= 2) {
             if (more == 0 && s->strstart == 0 && s->lookahead == 0) {
                 more = wsize;
 
             } else if (more == (unsigned)(-1)) {
                 /* Very unlikely, but possible on 16 bit machine if
                  * strstart == 0 && lookahead == 1 (input done a byte at time)
                  */
                 more--;
             }
         }
 
         /* If the window is almost full and there is insufficient lookahead,
          * move the upper half to the lower one to make room in the upper half.
          */
         if (s->strstart >= wsize+MAX_DIST(s)) {
 
-            zmemcpy(s->window, s->window+wsize, (unsigned)wsize);
+            zmemcpy(s->window, s->window+wsize, (unsigned)wsize - more);
             s->match_start -= wsize;
             s->strstart    -= wsize; /* we now have strstart >= MAX_DIST */
             s->block_start -= (long) wsize;
-
-            /* Slide the hash table (could be avoided with 32 bit values
-               at the expense of memory usage). We slide even when level == 0
-               to keep the hash table consistent if we switch back to level > 0
-               later. (Using level 0 permanently is not an optimal usage of
-               zlib, so we don't care about this pathological case.)
-             */
-            n = s->hash_size;
-            p = &s->head[n];
-            do {
-                m = *--p;
-                *p = (Pos)(m >= wsize ? m-wsize : NIL);
-            } while (--n);
-
-            n = wsize;
-#ifndef FASTEST
-            p = &s->prev[n];
-            do {
-                m = *--p;
-                *p = (Pos)(m >= wsize ? m-wsize : NIL);
-                /* If n is not on any hash chain, prev[n] is garbage but
-                 * its value will never be used.
-                 */
-            } while (--n);
-#endif
+            slide_hash(s);
             more += wsize;
         }
         if (s->strm->avail_in == 0) break;
 
         /* If there was no sliding:
          *    strstart <= WSIZE+MAX_DIST-1 && lookahead <= MIN_LOOKAHEAD - 1 &&
          *    more == window_size - lookahead - strstart
          * => more >= window_size - (MIN_LOOKAHEAD-1 + WSIZE + MAX_DIST-1)
          * => more >= window_size - 2*WSIZE + 2
          * In the BIG_MEM or MMAP case (not yet supported),
@@ skipping 85 matching lines @@
    flush_pending(s->strm); \
    Tracev((stderr,"[FLUSH]")); \
 }
 
 /* Same but force premature exit if necessary. */
 #define FLUSH_BLOCK(s, last) { \
    FLUSH_BLOCK_ONLY(s, last); \
    if (s->strm->avail_out == 0) return (last) ? finish_started : need_more; \
 }
 
+/* Maximum stored block length in deflate format (not including header). */
+#define MAX_STORED 65535
+
+/* Minimum of a and b. */
+#define MIN(a, b) ((a) > (b) ? (b) : (a))
+
 /* ===========================================================================
  * Copy without compression as much as possible from the input stream, return
  * the current block state.
- * This function does not insert new strings in the dictionary since
- * uncompressible data is probably not useful. This function is used
- * only for the level=0 compression option.
- * NOTE: this function should be optimized to avoid extra copying from
- * window to pending_buf.
+ *
+ * In case deflateParams() is used to later switch to a non-zero compression
+ * level, s->matches (otherwise unused when storing) keeps track of the number
+ * of hash table slides to perform. If s->matches is 1, then one hash table
+ * slide will be done when switching. If s->matches is 2, the maximum value
+ * allowed here, then the hash table will be cleared, since two or more slides
+ * is the same as a clear.
+ *
+ * deflate_stored() is written to minimize the number of times an input byte is
+ * copied. It is most efficient with large input and output buffers, which
+ * maximizes the opportunites to have a single copy from next_in to next_out.
  */
 local block_state deflate_stored(s, flush)
     deflate_state *s;
     int flush;
 {
-    /* Stored blocks are limited to 0xffff bytes, pending_buf is limited
-     * to pending_buf_size, and each stored block has a 5 byte header:
+    /* Smallest worthy block size when not flushing or finishing. By default
+     * this is 32K. This can be as small as 507 bytes for memLevel == 1. For
+     * large input and output buffers, the stored block size will be larger.
      */
-    ulg max_block_size = 0xffff;
-    ulg max_start;
+    unsigned min_block = MIN(s->pending_buf_size - 5, s->w_size);
 
-    if (max_block_size > s->pending_buf_size - 5) {
-        max_block_size = s->pending_buf_size - 5;
+    /* Copy as many min_block or larger stored blocks directly to next_out as
+     * possible. If flushing, copy the remaining available input to next_out as
+     * stored blocks, if there is enough space.
+     */
+    unsigned len, left, have, last = 0;
+    unsigned used = s->strm->avail_in;
+    do {
+        /* Set len to the maximum size block that we can copy directly with the
+         * available input data and output space. Set left to how much of that
+         * would be copied from what's left in the window.
+         */
+        len = MAX_STORED;       /* maximum deflate stored block length */
+        have = (s->bi_valid + 42) >> 3;         /* number of header bytes */
+        if (s->strm->avail_out < have)          /* need room for header */
+            break;
+            /* maximum stored block length that will fit in avail_out: */
+        have = s->strm->avail_out - have;
+        left = s->strstart - s->block_start;    /* bytes left in window */
+        if (len > (ulg)left + s->strm->avail_in)
+            len = left + s->strm->avail_in;     /* limit len to the input */
+        if (len > have)
+            len = have;                         /* limit len to the output */
+
+        /* If the stored block would be less than min_block in length, or if
+         * unable to copy all of the available input when flushing, then try
+         * copying to the window and the pending buffer instead. Also don't
+         * write an empty block when flushing -- deflate() does that.
+         */
+        if (len < min_block && ((len == 0 && flush != Z_FINISH) ||
+                                flush == Z_NO_FLUSH ||
+                                len != left + s->strm->avail_in))
+            break;
+
+        /* Make a dummy stored block in pending to get the header bytes,
+         * including any pending bits. This also updates the debugging counts.
+         */
+        last = flush == Z_FINISH && len == left + s->strm->avail_in ? 1 : 0;
+        _tr_stored_block(s, (char *)0, 0L, last);
+
+        /* Replace the lengths in the dummy stored block with len. */
+        s->pending_buf[s->pending - 4] = len;
+        s->pending_buf[s->pending - 3] = len >> 8;
+        s->pending_buf[s->pending - 2] = ~len;
+        s->pending_buf[s->pending - 1] = ~len >> 8;
+
+        /* Write the stored block header bytes. */
+        flush_pending(s->strm);
+
+#ifdef ZLIB_DEBUG
+        /* Update debugging counts for the data about to be copied. */
+        s->compressed_len += len << 3;
+        s->bits_sent += len << 3;
+#endif
+
+        /* Copy uncompressed bytes from the window to next_out. */
+        if (left) {
+            if (left > len)
+                left = len;
+            zmemcpy(s->strm->next_out, s->window + s->block_start, left);
+            s->strm->next_out += left;
+            s->strm->avail_out -= left;
+            s->strm->total_out += left;
+            s->block_start += left;
+            len -= left;
+        }
+
+        /* Copy uncompressed bytes directly from next_in to next_out, updating
+         * the check value.
+         */
+        if (len) {
+            read_buf(s->strm, s->strm->next_out, len);
+            s->strm->next_out += len;
+            s->strm->avail_out -= len;
+            s->strm->total_out += len;
+        }
+    } while (last == 0);
+
+    /* Update the sliding window with the last s->w_size bytes of the copied
+     * data, or append all of the copied data to the existing window if less
+     * than s->w_size bytes were copied. Also update the number of bytes to
+     * insert in the hash tables, in the event that deflateParams() switches to
+     * a non-zero compression level.
+     */
+    used -= s->strm->avail_in;      /* number of input bytes directly copied */
+    if (used) {
+        /* If any input was used, then no unused input remains in the window,
+         * therefore s->block_start == s->strstart.
+         */
+        if (used >= s->w_size) {    /* supplant the previous history */
+            s->matches = 2;         /* clear hash */
+            zmemcpy(s->window, s->strm->next_in - s->w_size, s->w_size);
+            s->strstart = s->w_size;
+        }
+        else {
+            if (s->window_size - s->strstart <= used) {
+                /* Slide the window down. */
+                s->strstart -= s->w_size;
+                zmemcpy(s->window, s->window + s->w_size, s->strstart);
+                if (s->matches < 2)
+                    s->matches++;   /* add a pending slide_hash() */
+            }
+            zmemcpy(s->window + s->strstart, s->strm->next_in - used, used);
+            s->strstart += used;
+        }
+        s->block_start = s->strstart;
+        s->insert += MIN(used, s->w_size - s->insert);
+    }
+    if (s->high_water < s->strstart)
+        s->high_water = s->strstart;
+
+    /* If the last block was written to next_out, then done. */
+    if (last)
+        return finish_done;
+
+    /* If flushing and all input has been consumed, then done. */
+    if (flush != Z_NO_FLUSH && flush != Z_FINISH &&
+        s->strm->avail_in == 0 && (long)s->strstart == s->block_start)
+        return block_done;
+
+    /* Fill the window with any remaining input. */
+    have = s->window_size - s->strstart - 1;
+    if (s->strm->avail_in > have && s->block_start >= (long)s->w_size) {
+        /* Slide the window down. */
+        s->block_start -= s->w_size;
+        s->strstart -= s->w_size;
+        zmemcpy(s->window, s->window + s->w_size, s->strstart);
+        if (s->matches < 2)
+            s->matches++;           /* add a pending slide_hash() */
+        have += s->w_size;          /* more space now */
+    }
+    if (have > s->strm->avail_in)
+        have = s->strm->avail_in;
+    if (have) {
+        read_buf(s->strm, s->window + s->strstart, have);
+        s->strstart += have;
+    }
+    if (s->high_water < s->strstart)
+        s->high_water = s->strstart;
+
+    /* There was not enough avail_out to write a complete worthy or flushed
+     * stored block to next_out. Write a stored block to pending instead, if we
+     * have enough input for a worthy block, or if flushing and there is enough
+     * room for the remaining input as a stored block in the pending buffer.
+     */
+    have = (s->bi_valid + 42) >> 3;         /* number of header bytes */
+        /* maximum stored block length that will fit in pending: */
+    have = MIN(s->pending_buf_size - have, MAX_STORED);
+    min_block = MIN(have, s->w_size);
+    left = s->strstart - s->block_start;
+    if (left >= min_block ||
+        ((left || flush == Z_FINISH) && flush != Z_NO_FLUSH &&
+         s->strm->avail_in == 0 && left <= have)) {
+        len = MIN(left, have);
+        last = flush == Z_FINISH && s->strm->avail_in == 0 &&
+               len == left ? 1 : 0;
+        _tr_stored_block(s, (charf *)s->window + s->block_start, len, last);
+        s->block_start += len;
+        flush_pending(s->strm);
     }
 
-    /* Copy as much as possible from input to output: */
-    for (;;) {
-        /* Fill the window as much as possible: */
-        if (s->lookahead <= 1) {
-
-            Assert(s->strstart < s->w_size+MAX_DIST(s) ||
-                   s->block_start >= (long)s->w_size, "slide too late");
-
-            fill_window(s);
-            if (s->lookahead == 0 && flush == Z_NO_FLUSH) return need_more;
-
-            if (s->lookahead == 0) break; /* flush the current block */
-        }
-        Assert(s->block_start >= 0L, "block gone");
-
-        s->strstart += s->lookahead;
-        s->lookahead = 0;
-
-        /* Emit a stored block if pending_buf will be full: */
-        max_start = s->block_start + max_block_size;
-        if (s->strstart == 0 || (ulg)s->strstart >= max_start) {
-            /* strstart == 0 is possible when wraparound on 16-bit machine */
-            s->lookahead = (uInt)(s->strstart - max_start);
-            s->strstart = (uInt)max_start;
-            FLUSH_BLOCK(s, 0);
-        }
-        /* Flush if we may have to slide, otherwise block_start may become
-         * negative and the data will be gone:
-         */
-        if (s->strstart - (uInt)s->block_start >= MAX_DIST(s)) {
-            FLUSH_BLOCK(s, 0);
-        }
-    }
-    s->insert = 0;
-    if (flush == Z_FINISH) {
-        FLUSH_BLOCK(s, 1);
-        return finish_done;
-    }
-    if ((long)s->strstart > s->block_start)
-        FLUSH_BLOCK(s, 0);
-    return block_done;
+    /* We've done all we can with the available input and output. */
+    return last ? finish_started : need_more;
 }
 
 /* ===========================================================================
  * Compress as much as possible from the input stream, return the current
  * block state.
  * This function does not perform lazy evaluation of matches and inserts
  * new strings in the dictionary only for unmatched strings or for short
  * matches. It is used only for the fast compression options.
  */
 local block_state deflate_fast(s, flush)
@@ skipping 256 matching lines @@
             scan = s->window + s->strstart - 1;
             prev = *scan;
             if (prev == *++scan && prev == *++scan && prev == *++scan) {
                 strend = s->window + s->strstart + MAX_MATCH;
                 do {
                 } while (prev == *++scan && prev == *++scan &&
                          prev == *++scan && prev == *++scan &&
                          prev == *++scan && prev == *++scan &&
                          prev == *++scan && prev == *++scan &&
                          scan < strend);
-                s->match_length = MAX_MATCH - (int)(strend - scan);
+                s->match_length = MAX_MATCH - (uInt)(strend - scan);
                 if (s->match_length > s->lookahead)
                     s->match_length = s->lookahead;
             }
             Assert(scan <= s->window+(uInt)(s->window_size-1), "wild scan");
         }
 
         /* Emit match if have run of MIN_MATCH or longer, else emit literal */
         if (s->match_length >= MIN_MATCH) {
             check_match(s, s->strstart, s->strstart - 1, s->match_length);
 
@@ skipping 86 matching lines @@
 #else
     /* This should never happen */
     assert(0);
 #endif
 
     ret = s->head[h & s->hash_mask];
     s->head[h & s->hash_mask] = str;
     s->prev[str & s->w_mask] = ret;
     return ret;
 }