[thirdparty/linux.git] / lib / lzo / lzo1x_compress.c

/*
 *  LZO1X Compressor from LZO
 *
 *  Copyright (C) 1996-2012 Markus F.X.J. Oberhumer <markus@oberhumer.com>
 *
 *  The full LZO package can be found at:
 *  http://www.oberhumer.com/opensource/lzo/
 *
 *  Changed for Linux kernel use by:
 *  Nitin Gupta <nitingupta910@gmail.com>
 *  Richard Purdie <rpurdie@openedhand.com>
 */

#include <linux/module.h>
#include <linux/kernel.h>
#include <asm/unaligned.h>
#include <linux/lzo.h>
#include "lzodefs.h"

static noinline size_t
lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
		    unsigned char *out, size_t *out_len,
		    size_t ti, void *wrkmem, signed char *state_offset,
		    const unsigned char bitstream_version)
{
	const unsigned char *ip;
	unsigned char *op;
	const unsigned char * const in_end = in + in_len;
	const unsigned char * const ip_end = in + in_len - 20;
	const unsigned char *ii;
	lzo_dict_t * const dict = (lzo_dict_t *) wrkmem;

	op = out;
	ip = in;
	ii = ip;
	ip += ti < 4 ? 4 - ti : 0;

	for (;;) {
		const unsigned char *m_pos = NULL;
		size_t t, m_len, m_off;
		u32 dv;
		u32 run_length = 0;
literal:
		ip += 1 + ((ip - ii) >> 5);
next:
		if (unlikely(ip >= ip_end))
			break;
		dv = get_unaligned_le32(ip);

		if (dv == 0 && bitstream_version) {
			const unsigned char *ir = ip + 4;
			const unsigned char *limit = ip_end
				< (ip + MAX_ZERO_RUN_LENGTH + 1)
				? ip_end : ip + MAX_ZERO_RUN_LENGTH + 1;
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && \
	defined(LZO_FAST_64BIT_MEMORY_ACCESS)
			u64 dv64;

			for (; (ir + 32) <= limit; ir += 32) {
				dv64 = get_unaligned((u64 *)ir);
				dv64 |= get_unaligned((u64 *)ir + 1);
				dv64 |= get_unaligned((u64 *)ir + 2);
				dv64 |= get_unaligned((u64 *)ir + 3);
				if (dv64)
					break;
			}
			for (; (ir + 8) <= limit; ir += 8) {
				dv64 = get_unaligned((u64 *)ir);
				if (dv64) {
#  if defined(__LITTLE_ENDIAN)
					ir += __builtin_ctzll(dv64) >> 3;
#  elif defined(__BIG_ENDIAN)
					ir += __builtin_clzll(dv64) >> 3;
#  else
#    error "missing endian definition"
#  endif
					break;
				}
			}
#else
			while ((ir < (const unsigned char *)
					ALIGN((uintptr_t)ir, 4)) &&
					(ir < limit) && (*ir == 0))
				ir++;
			for (; (ir + 4) <= limit; ir += 4) {
				dv = *((u32 *)ir);
				if (dv) {
#  if defined(__LITTLE_ENDIAN)
					ir += __builtin_ctz(dv) >> 3;
#  elif defined(__BIG_ENDIAN)
					ir += __builtin_clz(dv) >> 3;
#  else
#    error "missing endian definition"
#  endif
					break;
				}
			}
#endif
			while (likely(ir < limit) && unlikely(*ir == 0))
				ir++;
			run_length = ir - ip;
			if (run_length > MAX_ZERO_RUN_LENGTH)
				run_length = MAX_ZERO_RUN_LENGTH;
		} else {
			t = ((dv * 0x1824429d) >> (32 - D_BITS)) & D_MASK;
			m_pos = in + dict[t];
			dict[t] = (lzo_dict_t) (ip - in);
			if (unlikely(dv != get_unaligned_le32(m_pos)))
				goto literal;
		}

		ii -= ti;
		ti = 0;
		t = ip - ii;
		if (t != 0) {
			if (t <= 3) {
				op[*state_offset] |= t;
				COPY4(op, ii);
				op += t;
			} else if (t <= 16) {
				*op++ = (t - 3);
				COPY8(op, ii);
				COPY8(op + 8, ii + 8);
				op += t;
			} else {
				if (t <= 18) {
					*op++ = (t - 3);
				} else {
					size_t tt = t - 18;
					*op++ = 0;
					while (unlikely(tt > 255)) {
						tt -= 255;
						*op++ = 0;
					}
					*op++ = tt;
				}
				do {
					COPY8(op, ii);
					COPY8(op + 8, ii + 8);
					op += 16;
					ii += 16;
					t -= 16;
				} while (t >= 16);
				if (t > 0) do {
					*op++ = *ii++;
				} while (--t > 0);
			}
		}

		if (unlikely(run_length)) {
			ip += run_length;
			run_length -= MIN_ZERO_RUN_LENGTH;
			put_unaligned_le32((run_length << 21) | 0xfffc18
					   | (run_length & 0x7), op);
			op += 4;
			run_length = 0;
			*state_offset = -3;
			goto finished_writing_instruction;
		}

		m_len = 4;
		{
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && defined(LZO_USE_CTZ64)
		u64 v;
		v = get_unaligned((const u64 *) (ip + m_len)) ^
		    get_unaligned((const u64 *) (m_pos + m_len));
		if (unlikely(v == 0)) {
			do {
				m_len += 8;
				v = get_unaligned((const u64 *) (ip + m_len)) ^
				    get_unaligned((const u64 *) (m_pos + m_len));
				if (unlikely(ip + m_len >= ip_end))
					goto m_len_done;
			} while (v == 0);
		}
#  if defined(__LITTLE_ENDIAN)
		m_len += (unsigned) __builtin_ctzll(v) / 8;
#  elif defined(__BIG_ENDIAN)
		m_len += (unsigned) __builtin_clzll(v) / 8;
#  else
#    error "missing endian definition"
#  endif
#elif defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && defined(LZO_USE_CTZ32)
		u32 v;
		v = get_unaligned((const u32 *) (ip + m_len)) ^
		    get_unaligned((const u32 *) (m_pos + m_len));
		if (unlikely(v == 0)) {
			do {
				m_len += 4;
				v = get_unaligned((const u32 *) (ip + m_len)) ^
				    get_unaligned((const u32 *) (m_pos + m_len));
				if (v != 0)
					break;
				m_len += 4;
				v = get_unaligned((const u32 *) (ip + m_len)) ^
				    get_unaligned((const u32 *) (m_pos + m_len));
				if (unlikely(ip + m_len >= ip_end))
					goto m_len_done;
			} while (v == 0);
		}
#  if defined(__LITTLE_ENDIAN)
		m_len += (unsigned) __builtin_ctz(v) / 8;
#  elif defined(__BIG_ENDIAN)
		m_len += (unsigned) __builtin_clz(v) / 8;
#  else
#    error "missing endian definition"
#  endif
#else
		if (unlikely(ip[m_len] == m_pos[m_len])) {
			do {
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (ip[m_len] != m_pos[m_len])
					break;
				m_len += 1;
				if (unlikely(ip + m_len >= ip_end))
					goto m_len_done;
			} while (ip[m_len] == m_pos[m_len]);
		}
#endif
		}
m_len_done:

		m_off = ip - m_pos;
		ip += m_len;
		if (m_len <= M2_MAX_LEN && m_off <= M2_MAX_OFFSET) {
			m_off -= 1;
			*op++ = (((m_len - 1) << 5) | ((m_off & 7) << 2));
			*op++ = (m_off >> 3);
		} else if (m_off <= M3_MAX_OFFSET) {
			m_off -= 1;
			if (m_len <= M3_MAX_LEN)
				*op++ = (M3_MARKER | (m_len - 2));
			else {
				m_len -= M3_MAX_LEN;
				*op++ = M3_MARKER | 0;
				while (unlikely(m_len > 255)) {
					m_len -= 255;
					*op++ = 0;
				}
				*op++ = (m_len);
			}
			*op++ = (m_off << 2);
			*op++ = (m_off >> 6);
		} else {
			m_off -= 0x4000;
			if (m_len <= M4_MAX_LEN)
				*op++ = (M4_MARKER | ((m_off >> 11) & 8)
						| (m_len - 2));
			else {
				m_len -= M4_MAX_LEN;
				*op++ = (M4_MARKER | ((m_off >> 11) & 8));
				while (unlikely(m_len > 255)) {
					m_len -= 255;
					*op++ = 0;
				}
				*op++ = (m_len);
			}
			*op++ = (m_off << 2);
			*op++ = (m_off >> 6);
		}
		*state_offset = -2;
finished_writing_instruction:
		ii = ip;
		goto next;
	}
	*out_len = op - out;
	return in_end - (ii - ti);
}

int lzogeneric1x_1_compress(const unsigned char *in, size_t in_len,
		     unsigned char *out, size_t *out_len,
		     void *wrkmem, const unsigned char bitstream_version)
{
	const unsigned char *ip = in;
	unsigned char *op = out;
	unsigned char *data_start;
	size_t l = in_len;
	size_t t = 0;
	signed char state_offset = -2;
	unsigned int m4_max_offset;

	// LZO v0 will never write 17 as first byte (except for zero-length
	// input), so this is used to version the bitstream
	if (bitstream_version > 0) {
		*op++ = 17;
		*op++ = bitstream_version;
		m4_max_offset = M4_MAX_OFFSET_V1;
	} else {
		m4_max_offset = M4_MAX_OFFSET_V0;
	}

	data_start = op;

	while (l > 20) {
		size_t ll = l <= (m4_max_offset + 1) ? l : (m4_max_offset + 1);
		uintptr_t ll_end = (uintptr_t) ip + ll;
		if ((ll_end + ((t + ll) >> 5)) <= ll_end)
			break;
		BUILD_BUG_ON(D_SIZE * sizeof(lzo_dict_t) > LZO1X_1_MEM_COMPRESS);
		memset(wrkmem, 0, D_SIZE * sizeof(lzo_dict_t));
		t = lzo1x_1_do_compress(ip, ll, op, out_len, t, wrkmem,
					&state_offset, bitstream_version);
		ip += ll;
		op += *out_len;
		l  -= ll;
	}
	t += l;

	if (t > 0) {
		const unsigned char *ii = in + in_len - t;

		if (op == data_start && t <= 238) {
			*op++ = (17 + t);
		} else if (t <= 3) {
			op[state_offset] |= t;
		} else if (t <= 18) {
			*op++ = (t - 3);
		} else {
			size_t tt = t - 18;
			*op++ = 0;
			while (tt > 255) {
				tt -= 255;
				*op++ = 0;
			}
			*op++ = tt;
		}
		if (t >= 16) do {
			COPY8(op, ii);
			COPY8(op + 8, ii + 8);
			op += 16;
			ii += 16;
			t -= 16;
		} while (t >= 16);
		if (t > 0) do {
			*op++ = *ii++;
		} while (--t > 0);
	}

	*op++ = M4_MARKER | 1;
	*op++ = 0;
	*op++ = 0;

	*out_len = op - out;
	return LZO_E_OK;
}

int lzo1x_1_compress(const unsigned char *in, size_t in_len,
		     unsigned char *out, size_t *out_len,
		     void *wrkmem)
{
	return lzogeneric1x_1_compress(in, in_len, out, out_len, wrkmem, 0);
}

int lzorle1x_1_compress(const unsigned char *in, size_t in_len,
		     unsigned char *out, size_t *out_len,
		     void *wrkmem)
{
	return lzogeneric1x_1_compress(in, in_len, out, out_len,
				       wrkmem, LZO_VERSION);
}

EXPORT_SYMBOL_GPL(lzo1x_1_compress);
EXPORT_SYMBOL_GPL(lzorle1x_1_compress);

MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("LZO1X-1 Compressor");
Commit	Line	Data
64c70b1c	1	/*
8b975bd3	2	* LZO1X Compressor from LZO
64c70b1c	3	*
8b975bd3	4	* Copyright (C) 1996-2012 Markus F.X.J. Oberhumer <markus@oberhumer.com>
64c70b1c RP	5	*
	6	* The full LZO package can be found at:
	7	* http://www.oberhumer.com/opensource/lzo/
	8	*
8b975bd3	9	* Changed for Linux kernel use by:
64c70b1c RP	10	* Nitin Gupta <nitingupta910@gmail.com>
	11	* Richard Purdie <rpurdie@openedhand.com>
	12	*/
	13
	14	#include <linux/module.h>
	15	#include <linux/kernel.h>
64c70b1c	16	#include <asm/unaligned.h>
8b975bd3	17	#include <linux/lzo.h>
64c70b1c RP	18	#include "lzodefs.h"
	19
	20	static noinline size_t
8b975bd3 MO	21	lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
8b975bd3 MO	22	unsigned char out, size_t out_len,
45ec975e DR	23	size_t ti, void wrkmem, signed char state_offset,
45ec975e DR	24	const unsigned char bitstream_version)
64c70b1c	25	{
8b975bd3 MO	26	const unsigned char *ip;
8b975bd3 MO	27	unsigned char *op;
64c70b1c	28	const unsigned char * const in_end = in + in_len;
8b975bd3 MO	29	const unsigned char * const ip_end = in + in_len - 20;
	30	const unsigned char *ii;
	31	lzo_dict_t * const dict = (lzo_dict_t *) wrkmem;
64c70b1c	32
8b975bd3 MO	33	op = out;
	34	ip = in;
	35	ii = ip;
	36	ip += ti < 4 ? 4 - ti : 0;
64c70b1c RP	37
64c70b1c RP	38	for (;;) {
5ee4014a	39	const unsigned char *m_pos = NULL;
8b975bd3 MO	40	size_t t, m_len, m_off;
8b975bd3 MO	41	u32 dv;
5ee4014a	42	u32 run_length = 0;
64c70b1c	43	literal:
8b975bd3 MO	44	ip += 1 + ((ip - ii) >> 5);
8b975bd3 MO	45	next:
64c70b1c RP	46	if (unlikely(ip >= ip_end))
64c70b1c RP	47	break;
8b975bd3	48	dv = get_unaligned_le32(ip);
5ee4014a	49
45ec975e	50	if (dv == 0 && bitstream_version) {
5ee4014a DR	51	const unsigned char *ir = ip + 4;
	52	const unsigned char *limit = ip_end
	53	< (ip + MAX_ZERO_RUN_LENGTH + 1)
	54	? ip_end : ip + MAX_ZERO_RUN_LENGTH + 1;
	55	#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && \
	56	defined(LZO_FAST_64BIT_MEMORY_ACCESS)
	57	u64 dv64;
	58
	59	for (; (ir + 32) <= limit; ir += 32) {
	60	dv64 = get_unaligned((u64 *)ir);
	61	dv64 \|= get_unaligned((u64 *)ir + 1);
	62	dv64 \|= get_unaligned((u64 *)ir + 2);
	63	dv64 \|= get_unaligned((u64 *)ir + 3);
	64	if (dv64)
	65	break;
	66	}
	67	for (; (ir + 8) <= limit; ir += 8) {
	68	dv64 = get_unaligned((u64 *)ir);
	69	if (dv64) {
	70	# if defined(__LITTLE_ENDIAN)
	71	ir += __builtin_ctzll(dv64) >> 3;
	72	# elif defined(__BIG_ENDIAN)
	73	ir += __builtin_clzll(dv64) >> 3;
	74	# else
	75	# error "missing endian definition"
	76	# endif
	77	break;
	78	}
	79	}
	80	#else
	81	while ((ir < (const unsigned char *)
	82	ALIGN((uintptr_t)ir, 4)) &&
	83	(ir < limit) && (*ir == 0))
	84	ir++;
	85	for (; (ir + 4) <= limit; ir += 4) {
	86	dv = ((u32 )ir);
	87	if (dv) {
	88	# if defined(__LITTLE_ENDIAN)
	89	ir += __builtin_ctz(dv) >> 3;
	90	# elif defined(__BIG_ENDIAN)
	91	ir += __builtin_clz(dv) >> 3;
	92	# else
	93	# error "missing endian definition"
	94	# endif
	95	break;
	96	}
	97	}
	98	#endif
	99	while (likely(ir < limit) && unlikely(*ir == 0))
	100	ir++;
	101	run_length = ir - ip;
	102	if (run_length > MAX_ZERO_RUN_LENGTH)
	103	run_length = MAX_ZERO_RUN_LENGTH;
	104	} else {
	105	t = ((dv * 0x1824429d) >> (32 - D_BITS)) & D_MASK;
	106	m_pos = in + dict[t];
	107	dict[t] = (lzo_dict_t) (ip - in);
	108	if (unlikely(dv != get_unaligned_le32(m_pos)))
	109	goto literal;
	110	}
64c70b1c	111
8b975bd3 MO	112	ii -= ti;
	113	ti = 0;
	114	t = ip - ii;
	115	if (t != 0) {
64c70b1c	116	if (t <= 3) {
5ee4014a	117	op[*state_offset] \|= t;
8b975bd3 MO	118	COPY4(op, ii);
	119	op += t;
	120	} else if (t <= 16) {
64c70b1c	121	*op++ = (t - 3);
8b975bd3 MO	122	COPY8(op, ii);
	123	COPY8(op + 8, ii + 8);
	124	op += t;
64c70b1c	125	} else {
8b975bd3 MO	126	if (t <= 18) {
	127	*op++ = (t - 3);
	128	} else {
	129	size_t tt = t - 18;
64c70b1c	130	*op++ = 0;
8b975bd3 MO	131	while (unlikely(tt > 255)) {
	132	tt -= 255;
	133	*op++ = 0;
	134	}
	135	*op++ = tt;
64c70b1c	136	}
8b975bd3 MO	137	do {
	138	COPY8(op, ii);
	139	COPY8(op + 8, ii + 8);
	140	op += 16;
	141	ii += 16;
	142	t -= 16;
	143	} while (t >= 16);
	144	if (t > 0) do {
	145	op++ = ii++;
	146	} while (--t > 0);
64c70b1c	147	}
64c70b1c RP	148	}
64c70b1c RP	149
5ee4014a DR	150	if (unlikely(run_length)) {
	151	ip += run_length;
	152	run_length -= MIN_ZERO_RUN_LENGTH;
	153	put_unaligned_le32((run_length << 21) \| 0xfffc18
	154	\| (run_length & 0x7), op);
	155	op += 4;
	156	run_length = 0;
	157	*state_offset = -3;
	158	goto finished_writing_instruction;
	159	}
	160
8b975bd3 MO	161	m_len = 4;
	162	{
	163	#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && defined(LZO_USE_CTZ64)
	164	u64 v;
	165	v = get_unaligned((const u64 *) (ip + m_len)) ^
	166	get_unaligned((const u64 *) (m_pos + m_len));
	167	if (unlikely(v == 0)) {
	168	do {
	169	m_len += 8;
	170	v = get_unaligned((const u64 *) (ip + m_len)) ^
	171	get_unaligned((const u64 *) (m_pos + m_len));
	172	if (unlikely(ip + m_len >= ip_end))
	173	goto m_len_done;
	174	} while (v == 0);
	175	}
	176	# if defined(__LITTLE_ENDIAN)
	177	m_len += (unsigned) __builtin_ctzll(v) / 8;
	178	# elif defined(__BIG_ENDIAN)
	179	m_len += (unsigned) __builtin_clzll(v) / 8;
	180	# else
	181	# error "missing endian definition"
	182	# endif
	183	#elif defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && defined(LZO_USE_CTZ32)
	184	u32 v;
	185	v = get_unaligned((const u32 *) (ip + m_len)) ^
	186	get_unaligned((const u32 *) (m_pos + m_len));
	187	if (unlikely(v == 0)) {
	188	do {
	189	m_len += 4;
	190	v = get_unaligned((const u32 *) (ip + m_len)) ^
	191	get_unaligned((const u32 *) (m_pos + m_len));
	192	if (v != 0)
	193	break;
	194	m_len += 4;
	195	v = get_unaligned((const u32 *) (ip + m_len)) ^
	196	get_unaligned((const u32 *) (m_pos + m_len));
	197	if (unlikely(ip + m_len >= ip_end))
	198	goto m_len_done;
	199	} while (v == 0);
	200	}
	201	# if defined(__LITTLE_ENDIAN)
	202	m_len += (unsigned) __builtin_ctz(v) / 8;
	203	# elif defined(__BIG_ENDIAN)
	204	m_len += (unsigned) __builtin_clz(v) / 8;
	205	# else
	206	# error "missing endian definition"
	207	# endif
	208	#else
	209	if (unlikely(ip[m_len] == m_pos[m_len])) {
	210	do {
	211	m_len += 1;
	212	if (ip[m_len] != m_pos[m_len])
	213	break;
	214	m_len += 1;
	215	if (ip[m_len] != m_pos[m_len])
	216	break;
	217	m_len += 1;
	218	if (ip[m_len] != m_pos[m_len])
	219	break;
	220	m_len += 1;
	221	if (ip[m_len] != m_pos[m_len])
	222	break;
	223	m_len += 1;
	224	if (ip[m_len] != m_pos[m_len])
225	break;
226	m_len += 1;
227	if (ip[m_len] != m_pos[m_len])
228	break;
229	m_len += 1;
230	if (ip[m_len] != m_pos[m_len])
231	break;
232	m_len += 1;
233	if (unlikely(ip + m_len >= ip_end))
234	goto m_len_done;
235	} while (ip[m_len] == m_pos[m_len]);
236	}
237	#endif
238	}
239	m_len_done:
64c70b1c	240
8b975bd3 MO	241	m_off = ip - m_pos;
8b975bd3 MO	242	ip += m_len;
8b975bd3 MO	243	if (m_len <= M2_MAX_LEN && m_off <= M2_MAX_OFFSET) {
	244	m_off -= 1;
	245	*op++ = (((m_len - 1) << 5) \| ((m_off & 7) << 2));
	246	*op++ = (m_off >> 3);
	247	} else if (m_off <= M3_MAX_OFFSET) {
	248	m_off -= 1;
	249	if (m_len <= M3_MAX_LEN)
64c70b1c	250	*op++ = (M3_MARKER \| (m_len - 2));
8b975bd3 MO	251	else {
	252	m_len -= M3_MAX_LEN;
	253	*op++ = M3_MARKER \| 0;
	254	while (unlikely(m_len > 255)) {
	255	m_len -= 255;
	256	*op++ = 0;
	257	}
	258	*op++ = (m_len);
64c70b1c	259	}
8b975bd3 MO	260	*op++ = (m_off << 2);
8b975bd3 MO	261	*op++ = (m_off >> 6);
64c70b1c	262	} else {
8b975bd3 MO	263	m_off -= 0x4000;
	264	if (m_len <= M4_MAX_LEN)
	265	*op++ = (M4_MARKER \| ((m_off >> 11) & 8)
64c70b1c	266	\| (m_len - 2));
8b975bd3 MO	267	else {
	268	m_len -= M4_MAX_LEN;
	269	*op++ = (M4_MARKER \| ((m_off >> 11) & 8));
	270	while (unlikely(m_len > 255)) {
	271	m_len -= 255;
	272	*op++ = 0;
64c70b1c	273	}
8b975bd3	274	*op++ = (m_len);
64c70b1c	275	}
8b975bd3	276	*op++ = (m_off << 2);
64c70b1c RP	277	*op++ = (m_off >> 6);
64c70b1c RP	278	}
5ee4014a DR	279	*state_offset = -2;
	280	finished_writing_instruction:
	281	ii = ip;
8b975bd3	282	goto next;
64c70b1c	283	}
64c70b1c	284	*out_len = op - out;
8b975bd3	285	return in_end - (ii - ti);
64c70b1c RP	286	}
64c70b1c RP	287
45ec975e	288	int lzogeneric1x_1_compress(const unsigned char *in, size_t in_len,
8b975bd3	289	unsigned char out, size_t out_len,
45ec975e	290	void *wrkmem, const unsigned char bitstream_version)
64c70b1c	291	{
8b975bd3	292	const unsigned char *ip = in;
64c70b1c	293	unsigned char *op = out;
b11ed18e	294	unsigned char *data_start;
8b975bd3 MO	295	size_t l = in_len;
8b975bd3 MO	296	size_t t = 0;
5ee4014a	297	signed char state_offset = -2;
45ec975e	298	unsigned int m4_max_offset;
5ee4014a	299
b11ed18e DR	300	// LZO v0 will never write 17 as first byte (except for zero-length
b11ed18e DR	301	// input), so this is used to version the bitstream
45ec975e DR	302	if (bitstream_version > 0) {
	303	*op++ = 17;
	304	*op++ = bitstream_version;
	305	m4_max_offset = M4_MAX_OFFSET_V1;
	306	} else {
	307	m4_max_offset = M4_MAX_OFFSET_V0;
	308	}
64c70b1c	309
b11ed18e DR	310	data_start = op;
b11ed18e DR	311
8b975bd3	312	while (l > 20) {
45ec975e	313	size_t ll = l <= (m4_max_offset + 1) ? l : (m4_max_offset + 1);
8b975bd3 MO	314	uintptr_t ll_end = (uintptr_t) ip + ll;
	315	if ((ll_end + ((t + ll) >> 5)) <= ll_end)
	316	break;
	317	BUILD_BUG_ON(D_SIZE * sizeof(lzo_dict_t) > LZO1X_1_MEM_COMPRESS);
	318	memset(wrkmem, 0, D_SIZE * sizeof(lzo_dict_t));
45ec975e DR	319	t = lzo1x_1_do_compress(ip, ll, op, out_len, t, wrkmem,
45ec975e DR	320	&state_offset, bitstream_version);
8b975bd3	321	ip += ll;
64c70b1c	322	op += *out_len;
8b975bd3	323	l -= ll;
64c70b1c	324	}
8b975bd3	325	t += l;
64c70b1c RP	326
64c70b1c RP	327	if (t > 0) {
8b975bd3	328	const unsigned char *ii = in + in_len - t;
64c70b1c	329
b11ed18e	330	if (op == data_start && t <= 238) {
64c70b1c RP	331	*op++ = (17 + t);
64c70b1c RP	332	} else if (t <= 3) {
5ee4014a	333	op[state_offset] \|= t;
64c70b1c RP	334	} else if (t <= 18) {
	335	*op++ = (t - 3);
	336	} else {
	337	size_t tt = t - 18;
64c70b1c RP	338	*op++ = 0;
	339	while (tt > 255) {
	340	tt -= 255;
	341	*op++ = 0;
	342	}
64c70b1c RP	343	*op++ = tt;
64c70b1c RP	344	}
8b975bd3 MO	345	if (t >= 16) do {
	346	COPY8(op, ii);
	347	COPY8(op + 8, ii + 8);
	348	op += 16;
	349	ii += 16;
	350	t -= 16;
	351	} while (t >= 16);
	352	if (t > 0) do {
64c70b1c RP	353	op++ = ii++;
	354	} while (--t > 0);
	355	}
	356
	357	*op++ = M4_MARKER \| 1;
	358	*op++ = 0;
	359	*op++ = 0;
	360
	361	*out_len = op - out;
	362	return LZO_E_OK;
	363	}
45ec975e DR	364
	365	int lzo1x_1_compress(const unsigned char *in, size_t in_len,
	366	unsigned char out, size_t out_len,
	367	void *wrkmem)
	368	{
	369	return lzogeneric1x_1_compress(in, in_len, out, out_len, wrkmem, 0);
	370	}
	371
	372	int lzorle1x_1_compress(const unsigned char *in, size_t in_len,
	373	unsigned char out, size_t out_len,
	374	void *wrkmem)
	375	{
	376	return lzogeneric1x_1_compress(in, in_len, out, out_len,
	377	wrkmem, LZO_VERSION);
	378	}
	379
64c70b1c	380	EXPORT_SYMBOL_GPL(lzo1x_1_compress);
45ec975e	381	EXPORT_SYMBOL_GPL(lzorle1x_1_compress);
64c70b1c RP	382
	383	MODULE_LICENSE("GPL");
	384	MODULE_DESCRIPTION("LZO1X-1 Compressor");