[apple/xnu.git] / libkern / zlib / inffast.c

/*
 * Copyright (c) 2008 Apple Inc. All rights reserved.
 *
 * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
 * 
 * This file contains Original Code and/or Modifications of Original Code
 * as defined in and that are subject to the Apple Public Source License
 * Version 2.0 (the 'License'). You may not use this file except in
 * compliance with the License. The rights granted to you under the License
 * may not be used to create, or enable the creation or redistribution of,
 * unlawful or unlicensed copies of an Apple operating system, or to
 * circumvent, violate, or enable the circumvention or violation of, any
 * terms of an Apple operating system software license agreement.
 * 
 * Please obtain a copy of the License at
 * http://www.opensource.apple.com/apsl/ and read it before using this file.
 * 
 * The Original Code and all software distributed under the License are
 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
 * Please see the License for the specific language governing rights and
 * limitations under the License.
 * 
 * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
 */
/* inffast.c -- fast decoding
 * Copyright (C) 1995-2004 Mark Adler
 * For conditions of distribution and use, see copyright notice in zlib.h
 */

#include "zutil.h"
#include "inftrees.h"
#include "inflate.h"
#include "inffast.h"

#ifndef ASMINF

/* Allow machine dependent optimization for post-increment or pre-increment.
   Based on testing to date,
   Pre-increment preferred for:
   - PowerPC G3 (Adler)
   - MIPS R5000 (Randers-Pehrson)
   Post-increment preferred for:
   - none
   No measurable difference:
   - Pentium III (Anderson)
   - M68060 (Nikl)
 */
#ifdef POSTINC
#  define OFF 0
#  define PUP(a) *(a)++
#else
#  define OFF 1
#  define PUP(a) *++(a)
#endif

/*
   Decode literal, length, and distance codes and write out the resulting
   literal and match bytes until either not enough input or output is
   available, an end-of-block is encountered, or a data error is encountered.
   When large enough input and output buffers are supplied to inflate(), for
   example, a 16K input buffer and a 64K output buffer, more than 95% of the
   inflate execution time is spent in this routine.

   Entry assumptions:

        state->mode == LEN
        strm->avail_in >= 6
        strm->avail_out >= 258
        start >= strm->avail_out
        state->bits < 8

   On return, state->mode is one of:

        LEN -- ran out of enough output space or enough available input
        TYPE -- reached end of block code, inflate() to interpret next block
        BAD -- error in block data

   Notes:

    - The maximum input bits used by a length/distance pair is 15 bits for the
      length code, 5 bits for the length extra, 15 bits for the distance code,
      and 13 bits for the distance extra.  This totals 48 bits, or six bytes.
      Therefore if strm->avail_in >= 6, then there is enough input to avoid
      checking for available input while decoding.

    - The maximum bytes that a single length/distance pair can output is 258
      bytes, which is the maximum length that can be coded.  inflate_fast()
      requires strm->avail_out >= 258 for each loop to avoid checking for
      output space.
 */
void inflate_fast(strm, start)
z_streamp strm;
unsigned start;         /* inflate()'s starting value for strm->avail_out */
{
    struct inflate_state FAR *state;
    unsigned char FAR *in;      /* local strm->next_in */
    unsigned char FAR *last;    /* while in < last, enough input available */
    unsigned char FAR *out;     /* local strm->next_out */
    unsigned char FAR *beg;     /* inflate()'s initial strm->next_out */
    unsigned char FAR *end;     /* while out < end, enough space available */
#ifdef INFLATE_STRICT
    unsigned dmax;              /* maximum distance from zlib header */
#endif
    unsigned wsize;             /* window size or zero if not using window */
    unsigned whave;             /* valid bytes in the window */
    unsigned write;             /* window write index */
    unsigned char FAR *window;  /* allocated sliding window, if wsize != 0 */
    unsigned long hold;         /* local strm->hold */
    unsigned bits;              /* local strm->bits */
    code const FAR *lcode;      /* local strm->lencode */
    code const FAR *dcode;      /* local strm->distcode */
    unsigned lmask;             /* mask for first level of length codes */
    unsigned dmask;             /* mask for first level of distance codes */
    code this;                  /* retrieved table entry */
    unsigned op;                /* code bits, operation, extra bits, or */
                                /*  window position, window bytes to copy */
    unsigned len;               /* match length, unused bytes */
    unsigned dist;              /* match distance */
    unsigned char FAR *from;    /* where to copy match from */

    /* copy state to local variables */
    state = (struct inflate_state FAR *)strm->state;
    in = strm->next_in - OFF;
    last = in + (strm->avail_in - 5);
    out = strm->next_out - OFF;
    beg = out - (start - strm->avail_out);
    end = out + (strm->avail_out - 257);
#ifdef INFLATE_STRICT
    dmax = state->dmax;
#endif
    wsize = state->wsize;
    whave = state->whave;
    write = state->write;
    window = state->window;
    hold = state->hold;
    bits = state->bits;
    lcode = state->lencode;
    dcode = state->distcode;
    lmask = (1U << state->lenbits) - 1;
    dmask = (1U << state->distbits) - 1;

    /* decode literals and length/distances until end-of-block or not enough
       input data or output space */
    do {
        if (bits < 15) {
            hold += (unsigned long)(PUP(in)) << bits;
            bits += 8;
            hold += (unsigned long)(PUP(in)) << bits;
            bits += 8;
        }
        this = lcode[hold & lmask];
      dolen:
        op = (unsigned)(this.bits);
        hold >>= op;
        bits -= op;
        op = (unsigned)(this.op);
        if (op == 0) {                          /* literal */
            Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
                    "inflate:         literal '%c'\n" :
                    "inflate:         literal 0x%02x\n", this.val));
            PUP(out) = (unsigned char)(this.val);
        }
        else if (op & 16) {                     /* length base */
            len = (unsigned)(this.val);
            op &= 15;                           /* number of extra bits */
            if (op) {
                if (bits < op) {
                    hold += (unsigned long)(PUP(in)) << bits;
                    bits += 8;
                }
                len += (unsigned)hold & ((1U << op) - 1);
                hold >>= op;
                bits -= op;
            }
            Tracevv((stderr, "inflate:         length %u\n", len));
            if (bits < 15) {
                hold += (unsigned long)(PUP(in)) << bits;
                bits += 8;
                hold += (unsigned long)(PUP(in)) << bits;
                bits += 8;
            }
            this = dcode[hold & dmask];
          dodist:
            op = (unsigned)(this.bits);
            hold >>= op;
            bits -= op;
            op = (unsigned)(this.op);
            if (op & 16) {                      /* distance base */
                dist = (unsigned)(this.val);
                op &= 15;                       /* number of extra bits */
                if (bits < op) {
                    hold += (unsigned long)(PUP(in)) << bits;
                    bits += 8;
                    if (bits < op) {
                        hold += (unsigned long)(PUP(in)) << bits;
                        bits += 8;
                    }
                }
                dist += (unsigned)hold & ((1U << op) - 1);
#ifdef INFLATE_STRICT
                if (dist > dmax) {
                    strm->msg = (char *)"invalid distance too far back";
                    state->mode = BAD;
                    break;
                }
#endif
                hold >>= op;
                bits -= op;
                Tracevv((stderr, "inflate:         distance %u\n", dist));
                op = (unsigned)(out - beg);     /* max distance in output */
                if (dist > op) {                /* see if copy from window */
                    op = dist - op;             /* distance back in window */
                    if (op > whave) {
                        strm->msg = (char *)"invalid distance too far back";
                        state->mode = BAD;
                        break;
                    }
                    from = window - OFF;
                    if (write == 0) {           /* very common case */
                        from += wsize - op;
                        if (op < len) {         /* some from window */
                            len -= op;
                            do {
                                PUP(out) = PUP(from);
                            } while (--op);
                            from = out - dist;  /* rest from output */
                        }
                    }
                    else if (write < op) {      /* wrap around window */
                        from += wsize + write - op;
                        op -= write;
                        if (op < len) {         /* some from end of window */
                            len -= op;
                            do {
                                PUP(out) = PUP(from);
                            } while (--op);
                            from = window - OFF;
                            if (write < len) {  /* some from start of window */
                                op = write;
                                len -= op;
                                do {
                                    PUP(out) = PUP(from);
                                } while (--op);
                                from = out - dist;      /* rest from output */
                            }
                        }
                    }
                    else {                      /* contiguous in window */
                        from += write - op;
                        if (op < len) {         /* some from window */
                            len -= op;
                            do {
                                PUP(out) = PUP(from);
                            } while (--op);
                            from = out - dist;  /* rest from output */
                        }
                    }
                    while (len > 2) {
                        PUP(out) = PUP(from);
                        PUP(out) = PUP(from);
                        PUP(out) = PUP(from);
                        len -= 3;
                    }
                    if (len) {
                        PUP(out) = PUP(from);
                        if (len > 1)
                            PUP(out) = PUP(from);
                    }
                }
                else {
                    from = out - dist;          /* copy direct from output */
                    do {                        /* minimum length is three */
                        PUP(out) = PUP(from);
                        PUP(out) = PUP(from);
                        PUP(out) = PUP(from);
                        len -= 3;
                    } while (len > 2);
                    if (len) {
                        PUP(out) = PUP(from);
                        if (len > 1)
                            PUP(out) = PUP(from);
                    }
                }
            }
            else if ((op & 64) == 0) {          /* 2nd level distance code */
                this = dcode[this.val + (hold & ((1U << op) - 1))];
                goto dodist;
            }
            else {
                strm->msg = (char *)"invalid distance code";
                state->mode = BAD;
                break;
            }
        }
        else if ((op & 64) == 0) {              /* 2nd level length code */
            this = lcode[this.val + (hold & ((1U << op) - 1))];
            goto dolen;
        }
        else if (op & 32) {                     /* end-of-block */
            Tracevv((stderr, "inflate:         end of block\n"));
            state->mode = TYPE;
            break;
        }
        else {
            strm->msg = (char *)"invalid literal/length code";
            state->mode = BAD;
            break;
        }
    } while (in < last && out < end);

    /* return unused bytes (on entry, bits < 8, so in won't go too far back) */
    len = bits >> 3;
    in -= len;
    bits -= len << 3;
    hold &= (1U << bits) - 1;

    /* update state and return */
    strm->next_in = in + OFF;
    strm->next_out = out + OFF;
    strm->avail_in = (unsigned)(in < last ? 5 + (last - in) : 5 - (in - last));
    strm->avail_out = (unsigned)(out < end ?
                                 257 + (end - out) : 257 - (out - end));
    state->hold = hold;
    state->bits = bits;
    return;
}

/*
   inflate_fast() speedups that turned out slower (on a PowerPC G3 750CXe):
   - Using bit fields for code structure
   - Different op definition to avoid & for extra bits (do & for table bits)
   - Three separate decoding do-loops for direct, window, and write == 0
   - Special case for distance > 1 copies to do overlapped load and store copy
   - Explicit branch predictions (based on measured branch probabilities)
   - Deferring match copy and interspersed it with decoding subsequent codes
   - Swapping literal/length else
   - Swapping window/direct else
   - Larger unrolled copy loops (three is about right)
   - Moving len -= 3 statement into middle of loop
 */

#endif /* !ASMINF */
Commit	Line	Data
b0d623f7 A	1	/*
	2	* Copyright (c) 2008 Apple Inc. All rights reserved.
	3	*
	4	* @APPLE_OSREFERENCE_LICENSE_HEADER_START@
	5	*
	6	* This file contains Original Code and/or Modifications of Original Code
	7	* as defined in and that are subject to the Apple Public Source License
	8	* Version 2.0 (the 'License'). You may not use this file except in
	9	* compliance with the License. The rights granted to you under the License
	10	* may not be used to create, or enable the creation or redistribution of,
	11	* unlawful or unlicensed copies of an Apple operating system, or to
	12	* circumvent, violate, or enable the circumvention or violation of, any
	13	* terms of an Apple operating system software license agreement.
	14	*
	15	* Please obtain a copy of the License at
	16	* http://www.opensource.apple.com/apsl/ and read it before using this file.
	17	*
	18	* The Original Code and all software distributed under the License are
	19	* distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
	20	* EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
	21	* INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
	22	* FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
	23	* Please see the License for the specific language governing rights and
	24	* limitations under the License.
	25	*
	26	* @APPLE_OSREFERENCE_LICENSE_HEADER_END@
	27	*/
	28	/* inffast.c -- fast decoding
	29	* Copyright (C) 1995-2004 Mark Adler
	30	* For conditions of distribution and use, see copyright notice in zlib.h
	31	*/
	32
	33	#include "zutil.h"
	34	#include "inftrees.h"
	35	#include "inflate.h"
	36	#include "inffast.h"
	37
	38	#ifndef ASMINF
	39
	40	/* Allow machine dependent optimization for post-increment or pre-increment.
	41	Based on testing to date,
	42	Pre-increment preferred for:
	43	- PowerPC G3 (Adler)
	44	- MIPS R5000 (Randers-Pehrson)
	45	Post-increment preferred for:
	46	- none
	47	No measurable difference:
	48	- Pentium III (Anderson)
	49	- M68060 (Nikl)
	50	*/
	51	#ifdef POSTINC
	52	# define OFF 0
	53	# define PUP(a) *(a)++
	54	#else
	55	# define OFF 1
	56	# define PUP(a) *++(a)
	57	#endif
	58
	59	/*
	60	Decode literal, length, and distance codes and write out the resulting
	61	literal and match bytes until either not enough input or output is
	62	available, an end-of-block is encountered, or a data error is encountered.
	63	When large enough input and output buffers are supplied to inflate(), for
	64	example, a 16K input buffer and a 64K output buffer, more than 95% of the
65	inflate execution time is spent in this routine.
66
67	Entry assumptions:
68
69	state->mode == LEN
70	strm->avail_in >= 6
71	strm->avail_out >= 258
72	start >= strm->avail_out
73	state->bits < 8
74
75	On return, state->mode is one of:
76
77	LEN -- ran out of enough output space or enough available input
78	TYPE -- reached end of block code, inflate() to interpret next block
79	BAD -- error in block data
80
81	Notes:
82
83	- The maximum input bits used by a length/distance pair is 15 bits for the
84	length code, 5 bits for the length extra, 15 bits for the distance code,
85	and 13 bits for the distance extra. This totals 48 bits, or six bytes.
86	Therefore if strm->avail_in >= 6, then there is enough input to avoid
87	checking for available input while decoding.
88
89	- The maximum bytes that a single length/distance pair can output is 258
90	bytes, which is the maximum length that can be coded. inflate_fast()
91	requires strm->avail_out >= 258 for each loop to avoid checking for
92	output space.
93	*/
94	void inflate_fast(strm, start)
95	z_streamp strm;
96	unsigned start; /* inflate()'s starting value for strm->avail_out */
97	{
98	struct inflate_state FAR *state;
99	unsigned char FAR in; / local strm->next_in */
100	unsigned char FAR last; / while in < last, enough input available */
101	unsigned char FAR out; / local strm->next_out */
102	unsigned char FAR beg; / inflate()'s initial strm->next_out */
103	unsigned char FAR end; / while out < end, enough space available */
104	#ifdef INFLATE_STRICT
105	unsigned dmax; /* maximum distance from zlib header */
106	#endif
107	unsigned wsize; /* window size or zero if not using window */
108	unsigned whave; /* valid bytes in the window */
109	unsigned write; /* window write index */
110	unsigned char FAR window; / allocated sliding window, if wsize != 0 */
111	unsigned long hold; /* local strm->hold */
112	unsigned bits; /* local strm->bits */
113	code const FAR lcode; / local strm->lencode */
114	code const FAR dcode; / local strm->distcode */
115	unsigned lmask; /* mask for first level of length codes */
116	unsigned dmask; /* mask for first level of distance codes */
117	code this; /* retrieved table entry */
118	unsigned op; /* code bits, operation, extra bits, or */
119	/* window position, window bytes to copy */
120	unsigned len; /* match length, unused bytes */
121	unsigned dist; /* match distance */
122	unsigned char FAR from; / where to copy match from */
123
124	/* copy state to local variables */
125	state = (struct inflate_state FAR *)strm->state;
126	in = strm->next_in - OFF;
127	last = in + (strm->avail_in - 5);
128	out = strm->next_out - OFF;
129	beg = out - (start - strm->avail_out);
130	end = out + (strm->avail_out - 257);
131	#ifdef INFLATE_STRICT
132	dmax = state->dmax;
133	#endif
134	wsize = state->wsize;
135	whave = state->whave;
136	write = state->write;
137	window = state->window;
138	hold = state->hold;
139	bits = state->bits;
140	lcode = state->lencode;
141	dcode = state->distcode;
142	lmask = (1U << state->lenbits) - 1;
143	dmask = (1U << state->distbits) - 1;
144
145	/* decode literals and length/distances until end-of-block or not enough
146	input data or output space */
147	do {
148	if (bits < 15) {
149	hold += (unsigned long)(PUP(in)) << bits;
150	bits += 8;
151	hold += (unsigned long)(PUP(in)) << bits;
152	bits += 8;
153	}
154	this = lcode[hold & lmask];
155	dolen:
156	op = (unsigned)(this.bits);
157	hold >>= op;
158	bits -= op;
159	op = (unsigned)(this.op);
160	if (op == 0) { /* literal */
161	Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
162	"inflate: literal '%c'\n" :
163	"inflate: literal 0x%02x\n", this.val));
164	PUP(out) = (unsigned char)(this.val);
165	}
166	else if (op & 16) { /* length base */
167	len = (unsigned)(this.val);
168	op &= 15; /* number of extra bits */
169	if (op) {
170	if (bits < op) {
171	hold += (unsigned long)(PUP(in)) << bits;
172	bits += 8;
173	}
174	len += (unsigned)hold & ((1U << op) - 1);
175	hold >>= op;
176	bits -= op;
177	}
178	Tracevv((stderr, "inflate: length %u\n", len));
179	if (bits < 15) {
180	hold += (unsigned long)(PUP(in)) << bits;
181	bits += 8;
182	hold += (unsigned long)(PUP(in)) << bits;
183	bits += 8;
184	}
185	this = dcode[hold & dmask];
186	dodist:
187	op = (unsigned)(this.bits);
188	hold >>= op;
189	bits -= op;
190	op = (unsigned)(this.op);
191	if (op & 16) { /* distance base */
192	dist = (unsigned)(this.val);
193	op &= 15; /* number of extra bits */
194	if (bits < op) {
195	hold += (unsigned long)(PUP(in)) << bits;
196	bits += 8;
197	if (bits < op) {
198	hold += (unsigned long)(PUP(in)) << bits;
199	bits += 8;
200	}
201	}
202	dist += (unsigned)hold & ((1U << op) - 1);
203	#ifdef INFLATE_STRICT
204	if (dist > dmax) {
205	strm->msg = (char *)"invalid distance too far back";
206	state->mode = BAD;
207	break;
208	}
209	#endif
210	hold >>= op;
211	bits -= op;
212	Tracevv((stderr, "inflate: distance %u\n", dist));
213	op = (unsigned)(out - beg); /* max distance in output */
214	if (dist > op) { /* see if copy from window */
215	op = dist - op; /* distance back in window */
216	if (op > whave) {
217	strm->msg = (char *)"invalid distance too far back";
218	state->mode = BAD;
219	break;
220	}
221	from = window - OFF;
222	if (write == 0) { /* very common case */
223	from += wsize - op;
224	if (op < len) { /* some from window */
225	len -= op;
226	do {
227	PUP(out) = PUP(from);
228	} while (--op);
229	from = out - dist; /* rest from output */
230	}
231	}
232	else if (write < op) { /* wrap around window */
233	from += wsize + write - op;
234	op -= write;
235	if (op < len) { /* some from end of window */
236	len -= op;
237	do {
238	PUP(out) = PUP(from);
239	} while (--op);
240	from = window - OFF;
241	if (write < len) { /* some from start of window */
242	op = write;
243	len -= op;
244	do {
245	PUP(out) = PUP(from);
246	} while (--op);
247	from = out - dist; /* rest from output */
248	}
249	}
250	}
251	else { /* contiguous in window */
252	from += write - op;
253	if (op < len) { /* some from window */
254	len -= op;
255	do {
256	PUP(out) = PUP(from);
257	} while (--op);
258	from = out - dist; /* rest from output */
259	}
260	}
261	while (len > 2) {
262	PUP(out) = PUP(from);
263	PUP(out) = PUP(from);
264	PUP(out) = PUP(from);
265	len -= 3;
266	}
267	if (len) {
268	PUP(out) = PUP(from);
269	if (len > 1)
270	PUP(out) = PUP(from);
271	}
272	}
273	else {
274	from = out - dist; /* copy direct from output */
275	do { /* minimum length is three */
276	PUP(out) = PUP(from);
277	PUP(out) = PUP(from);
278	PUP(out) = PUP(from);
279	len -= 3;
280	} while (len > 2);
281	if (len) {
282	PUP(out) = PUP(from);
283	if (len > 1)
284	PUP(out) = PUP(from);
285	}
286	}
287	}
288	else if ((op & 64) == 0) { /* 2nd level distance code */
289	this = dcode[this.val + (hold & ((1U << op) - 1))];
290	goto dodist;
291	}
292	else {
293	strm->msg = (char *)"invalid distance code";
294	state->mode = BAD;
295	break;
296	}
297	}
298	else if ((op & 64) == 0) { /* 2nd level length code */
299	this = lcode[this.val + (hold & ((1U << op) - 1))];
300	goto dolen;
301	}
302	else if (op & 32) { /* end-of-block */
303	Tracevv((stderr, "inflate: end of block\n"));
304	state->mode = TYPE;
305	break;
306	}
307	else {
308	strm->msg = (char *)"invalid literal/length code";
309	state->mode = BAD;
310	break;
311	}
312	} while (in < last && out < end);
313
314	/* return unused bytes (on entry, bits < 8, so in won't go too far back) */
315	len = bits >> 3;
316	in -= len;
317	bits -= len << 3;
318	hold &= (1U << bits) - 1;
319
320	/* update state and return */
321	strm->next_in = in + OFF;
322	strm->next_out = out + OFF;
323	strm->avail_in = (unsigned)(in < last ? 5 + (last - in) : 5 - (in - last));
324	strm->avail_out = (unsigned)(out < end ?
325	257 + (end - out) : 257 - (out - end));
326	state->hold = hold;
327	state->bits = bits;
328	return;
329	}
330
331	/*
332	inflate_fast() speedups that turned out slower (on a PowerPC G3 750CXe):
333	- Using bit fields for code structure
334	- Different op definition to avoid & for extra bits (do & for table bits)
335	- Three separate decoding do-loops for direct, window, and write == 0
336	- Special case for distance > 1 copies to do overlapped load and store copy
337	- Explicit branch predictions (based on measured branch probabilities)
338	- Deferring match copy and interspersed it with decoding subsequent codes
339	- Swapping literal/length else
340	- Swapping window/direct else
341	- Larger unrolled copy loops (three is about right)
342	- Moving len -= 3 statement into middle of loop
343	*/
344
345	#endif /* !ASMINF */