Merge branch 'master' of /repos/git/net-next-2.6

Signed-off-by: Patrick McHardy <kaber@trash.net>

11 files changed, 476 insertions, 26 deletions

diff --git a/lib/Kconfig b/lib/Kconfig
index 1cfe51628e1..97b136ff117 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -117,6 +117,10 @@ config DECOMPRESS_BZIP2
 config DECOMPRESS_LZMA
 	tristate
 
+config DECOMPRESS_LZO
+	select LZO_DECOMPRESS
+	tristate
+
 #
 # Generic allocator support is selected if needed
 #
diff --git a/lib/Makefile b/lib/Makefile
index 347ad8db29d..3b0b4a696db 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -21,7 +21,7 @@ lib-y	+= kobject.o kref.o klist.o
 
 obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \
 	 bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o \
-	 string_helpers.o gcd.o
+	 string_helpers.o gcd.o list_sort.o
 
 ifeq ($(CONFIG_DEBUG_KOBJECT),y)
 CFLAGS_kobject.o += -DDEBUG
@@ -69,6 +69,7 @@ obj-$(CONFIG_LZO_DECOMPRESS) += lzo/
 lib-$(CONFIG_DECOMPRESS_GZIP) += decompress_inflate.o
 lib-$(CONFIG_DECOMPRESS_BZIP2) += decompress_bunzip2.o
 lib-$(CONFIG_DECOMPRESS_LZMA) += decompress_unlzma.o
+lib-$(CONFIG_DECOMPRESS_LZO) += decompress_unlzo.o
 
 obj-$(CONFIG_TEXTSEARCH) += textsearch.o
 obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
diff --git a/lib/decompress.c b/lib/decompress.c
index d2842f57167..a7606815541 100644
--- a/lib/decompress.c
+++ b/lib/decompress.c
@@ -9,6 +9,7 @@
 #include <linux/decompress/bunzip2.h>
 #include <linux/decompress/unlzma.h>
 #include <linux/decompress/inflate.h>
+#include <linux/decompress/unlzo.h>
 
 #include <linux/types.h>
 #include <linux/string.h>
@@ -22,6 +23,9 @@
 #ifndef CONFIG_DECOMPRESS_LZMA
 # define unlzma NULL
 #endif
+#ifndef CONFIG_DECOMPRESS_LZO
+# define unlzo NULL
+#endif
 
 static const struct compress_format {
 	unsigned char magic[2];
@@ -32,6 +36,7 @@ static const struct compress_format {
 	{ {037, 0236}, "gzip", gunzip },
 	{ {0x42, 0x5a}, "bzip2", bunzip2 },
 	{ {0x5d, 0x00}, "lzma", unlzma },
+	{ {0x89, 0x4c}, "lzo", unlzo },
 	{ {0, 0}, NULL, NULL }
 };
 
diff --git a/lib/decompress_unlzo.c b/lib/decompress_unlzo.c
new file mode 100644
index 00000000000..db521f45626
--- /dev/null
+++ b/lib/decompress_unlzo.c
@@ -0,0 +1,209 @@
+/*
+ * LZO decompressor for the Linux kernel. Code borrowed from the lzo
+ * implementation by Markus Franz Xaver Johannes Oberhumer.
+ *
+ * Linux kernel adaptation:
+ * Copyright (C) 2009
+ * Albin Tonnerre, Free Electrons <albin.tonnerre@free-electrons.com>
+ *
+ * Original code:
+ * Copyright (C) 1996-2005 Markus Franz Xaver Johannes Oberhumer
+ * All Rights Reserved.
+ *
+ * lzop and the LZO library are free software; you can redistribute them
+ * and/or modify them under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING.
+ * If not, write to the Free Software Foundation, Inc.,
+ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Markus F.X.J. Oberhumer
+ * <markus@oberhumer.com>
+ * http://www.oberhumer.com/opensource/lzop/
+ */
+
+#ifdef STATIC
+#include "lzo/lzo1x_decompress.c"
+#else
+#include <linux/slab.h>
+#include <linux/decompress/unlzo.h>
+#endif
+
+#include <linux/types.h>
+#include <linux/lzo.h>
+#include <linux/decompress/mm.h>
+
+#include <linux/compiler.h>
+#include <asm/unaligned.h>
+
+static const unsigned char lzop_magic[] = {
+	0x89, 0x4c, 0x5a, 0x4f, 0x00, 0x0d, 0x0a, 0x1a, 0x0a };
+
+#define LZO_BLOCK_SIZE        (256*1024l)
+#define HEADER_HAS_FILTER      0x00000800L
+
+STATIC inline int INIT parse_header(u8 *input, u8 *skip)
+{
+	int l;
+	u8 *parse = input;
+	u8 level = 0;
+	u16 version;
+
+	/* read magic: 9 first bits */
+	for (l = 0; l < 9; l++) {
+		if (*parse++ != lzop_magic[l])
+			return 0;
+	}
+	/* get version (2bytes), skip library version (2),
+	 * 'need to be extracted' version (2) and
+	 * method (1) */
+	version = get_unaligned_be16(parse);
+	parse += 7;
+	if (version >= 0x0940)
+		level = *parse++;
+	if (get_unaligned_be32(parse) & HEADER_HAS_FILTER)
+		parse += 8; /* flags + filter info */
+	else
+		parse += 4; /* flags */
+
+	/* skip mode and mtime_low */
+	parse += 8;
+	if (version >= 0x0940)
+		parse += 4;	/* skip mtime_high */
+
+	l = *parse++;
+	/* don't care about the file name, and skip checksum */
+	parse += l + 4;
+
+	*skip = parse - input;
+	return 1;
+}
+
+STATIC inline int INIT unlzo(u8 *input, int in_len,
+				int (*fill) (void *, unsigned int),
+				int (*flush) (void *, unsigned int),
+				u8 *output, int *posp,
+				void (*error_fn) (char *x))
+{
+	u8 skip = 0, r = 0;
+	u32 src_len, dst_len;
+	size_t tmp;
+	u8 *in_buf, *in_buf_save, *out_buf;
+	int obytes_processed = 0;
+
+	set_error_fn(error_fn);
+
+	if (output) {
+		out_buf = output;
+	} else if (!flush) {
+		error("NULL output pointer and no flush function provided");
+		goto exit;
+	} else {
+		out_buf = malloc(LZO_BLOCK_SIZE);
+		if (!out_buf) {
+			error("Could not allocate output buffer");
+			goto exit;
+		}
+	}
+
+	if (input && fill) {
+		error("Both input pointer and fill function provided, don't know what to do");
+		goto exit_1;
+	} else if (input) {
+		in_buf = input;
+	} else if (!fill || !posp) {
+		error("NULL input pointer and missing position pointer or fill function");
+		goto exit_1;
+	} else {
+		in_buf = malloc(lzo1x_worst_compress(LZO_BLOCK_SIZE));
+		if (!in_buf) {
+			error("Could not allocate input buffer");
+			goto exit_1;
+		}
+	}
+	in_buf_save = in_buf;
+
+	if (posp)
+		*posp = 0;
+
+	if (fill)
+		fill(in_buf, lzo1x_worst_compress(LZO_BLOCK_SIZE));
+
+	if (!parse_header(input, &skip)) {
+		error("invalid header");
+		goto exit_2;
+	}
+	in_buf += skip;
+
+	if (posp)
+		*posp = skip;
+
+	for (;;) {
+		/* read uncompressed block size */
+		dst_len = get_unaligned_be32(in_buf);
+		in_buf += 4;
+
+		/* exit if last block */
+		if (dst_len == 0) {
+			if (posp)
+				*posp += 4;
+			break;
+		}
+
+		if (dst_len > LZO_BLOCK_SIZE) {
+			error("dest len longer than block size");
+			goto exit_2;
+		}
+
+		/* read compressed block size, and skip block checksum info */
+		src_len = get_unaligned_be32(in_buf);
+		in_buf += 8;
+
+		if (src_len <= 0 || src_len > dst_len) {
+			error("file corrupted");
+			goto exit_2;
+		}
+
+		/* decompress */
+		tmp = dst_len;
+		r = lzo1x_decompress_safe((u8 *) in_buf, src_len,
+						out_buf, &tmp);
+
+		if (r != LZO_E_OK || dst_len != tmp) {
+			error("Compressed data violation");
+			goto exit_2;
+		}
+
+		obytes_processed += dst_len;
+		if (flush)
+			flush(out_buf, dst_len);
+		if (output)
+			out_buf += dst_len;
+		if (posp)
+			*posp += src_len + 12;
+		if (fill) {
+			in_buf = in_buf_save;
+			fill(in_buf, lzo1x_worst_compress(LZO_BLOCK_SIZE));
+		} else
+			in_buf += src_len;
+	}
+
+exit_2:
+	if (!input)
+		free(in_buf);
+exit_1:
+	if (!output)
+		free(out_buf);
+exit:
+	return obytes_processed;
+}
+
+#define decompress unlzo
diff --git a/lib/dma-debug.c b/lib/dma-debug.c
index d9b08e0f7f5..7d2f0b33e5a 100644
--- a/lib/dma-debug.c
+++ b/lib/dma-debug.c
@@ -670,12 +670,13 @@ static int device_dma_allocations(struct device *dev)
 	return count;
 }
 
-static int dma_debug_device_change(struct notifier_block *nb,
-				    unsigned long action, void *data)
+static int dma_debug_device_change(struct notifier_block *nb, unsigned long action, void *data)
 {
 	struct device *dev = data;
 	int count;
 
+	if (global_disable)
+		return 0;
 
 	switch (action) {
 	case BUS_NOTIFY_UNBOUND_DRIVER:
@@ -697,6 +698,9 @@ void dma_debug_add_bus(struct bus_type *bus)
 {
 	struct notifier_block *nb;
 
+	if (global_disable)
+		return;
+
 	nb = kzalloc(sizeof(struct notifier_block), GFP_KERNEL);
 	if (nb == NULL) {
 		pr_err("dma_debug_add_bus: out of memory\n");
@@ -909,6 +913,9 @@ static void check_sync(struct device *dev,
 				ref->size);
 	}
 
+	if (entry->direction == DMA_BIDIRECTIONAL)
+		goto out;
+
 	if (ref->direction != entry->direction) {
 		err_printk(dev, entry, "DMA-API: device driver syncs "
 				"DMA memory with different direction "
@@ -919,9 +926,6 @@ static void check_sync(struct device *dev,
 				dir2name[ref->direction]);
 	}
 
-	if (entry->direction == DMA_BIDIRECTIONAL)
-		goto out;
-
 	if (to_cpu && !(entry->direction == DMA_FROM_DEVICE) &&
 		      !(ref->direction == DMA_TO_DEVICE))
 		err_printk(dev, entry, "DMA-API: device driver syncs "
@@ -944,7 +948,6 @@ static void check_sync(struct device *dev,
 
 out:
 	put_hash_bucket(bucket, &flags);
-
 }
 
 void debug_dma_map_page(struct device *dev, struct page *page, size_t offset,
diff --git a/lib/list_sort.c b/lib/list_sort.c
new file mode 100644
index 00000000000..19d11e0bb95
--- /dev/null
+++ b/lib/list_sort.c
@@ -0,0 +1,102 @@
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/list_sort.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+
+/**
+ * list_sort - sort a list.
+ * @priv: private data, passed to @cmp
+ * @head: the list to sort
+ * @cmp: the elements comparison function
+ *
+ * This function has been implemented by Mark J Roberts <mjr@znex.org>. It
+ * implements "merge sort" which has O(nlog(n)) complexity. The list is sorted
+ * in ascending order.
+ *
+ * The comparison function @cmp is supposed to return a negative value if @a is
+ * less than @b, and a positive value if @a is greater than @b. If @a and @b
+ * are equivalent, then it does not matter what this function returns.
+ */
+void list_sort(void *priv, struct list_head *head,
+	       int (*cmp)(void *priv, struct list_head *a,
+			  struct list_head *b))
+{
+	struct list_head *p, *q, *e, *list, *tail, *oldhead;
+	int insize, nmerges, psize, qsize, i;
+
+	if (list_empty(head))
+		return;
+
+	list = head->next;
+	list_del(head);
+	insize = 1;
+	for (;;) {
+		p = oldhead = list;
+		list = tail = NULL;
+		nmerges = 0;
+
+		while (p) {
+			nmerges++;
+			q = p;
+			psize = 0;
+			for (i = 0; i < insize; i++) {
+				psize++;
+				q = q->next == oldhead ? NULL : q->next;
+				if (!q)
+					break;
+			}
+
+			qsize = insize;
+			while (psize > 0 || (qsize > 0 && q)) {
+				if (!psize) {
+					e = q;
+					q = q->next;
+					qsize--;
+					if (q == oldhead)
+						q = NULL;
+				} else if (!qsize || !q) {
+					e = p;
+					p = p->next;
+					psize--;
+					if (p == oldhead)
+						p = NULL;
+				} else if (cmp(priv, p, q) <= 0) {
+					e = p;
+					p = p->next;
+					psize--;
+					if (p == oldhead)
+						p = NULL;
+				} else {
+					e = q;
+					q = q->next;
+					qsize--;
+					if (q == oldhead)
+						q = NULL;
+				}
+				if (tail)
+					tail->next = e;
+				else
+					list = e;
+				e->prev = tail;
+				tail = e;
+			}
+			p = q;
+		}
+
+		tail->next = list;
+		list->prev = tail;
+
+		if (nmerges <= 1)
+			break;
+
+		insize *= 2;
+	}
+
+	head->next = list;
+	head->prev = list->prev;
+	list->prev->next = head;
+	list->prev = head;
+}
+
+EXPORT_SYMBOL(list_sort);
diff --git a/lib/lzo/lzo1x_decompress.c b/lib/lzo/lzo1x_decompress.c
index 5dc6b29c157..f2fd0985022 100644
--- a/lib/lzo/lzo1x_decompress.c
+++ b/lib/lzo/lzo1x_decompress.c
@@ -11,11 +11,13 @@
  *  Richard Purdie <rpurdie@openedhand.com>
  */
 
+#ifndef STATIC
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/lzo.h>
-#include <asm/byteorder.h>
+#endif
+
 #include <asm/unaligned.h>
+#include <linux/lzo.h>
 #include "lzodefs.h"
 
 #define HAVE_IP(x, ip_end, ip) ((size_t)(ip_end - ip) < (x))
@@ -244,9 +246,10 @@ lookbehind_overrun:
 	*out_len = op - out;
 	return LZO_E_LOOKBEHIND_OVERRUN;
 }
-
+#ifndef STATIC
 EXPORT_SYMBOL_GPL(lzo1x_decompress_safe);
 
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("LZO1X Decompressor");
 
+#endif
diff --git a/lib/rational.c b/lib/rational.c
index b3c099b5478..3ed247b8066 100644
--- a/lib/rational.c
+++ b/lib/rational.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/rational.h>
+#include <linux/module.h>
 
 /*
  * calculate best rational approximation for a given fraction
diff --git a/lib/string.c b/lib/string.c
index 9f75b4ec50b..a1cdcfcc42d 100644
--- a/lib/string.c
+++ b/lib/string.c
@@ -667,7 +667,7 @@ EXPORT_SYMBOL(memscan);
  */
 char *strstr(const char *s1, const char *s2)
 {
-	int l1, l2;
+	size_t l1, l2;
 
 	l2 = strlen(s2);
 	if (!l2)
@@ -684,6 +684,31 @@ char *strstr(const char *s1, const char *s2)
 EXPORT_SYMBOL(strstr);
 #endif
 
+#ifndef __HAVE_ARCH_STRNSTR
+/**
+ * strnstr - Find the first substring in a length-limited string
+ * @s1: The string to be searched
+ * @s2: The string to search for
+ * @len: the maximum number of characters to search
+ */
+char *strnstr(const char *s1, const char *s2, size_t len)
+{
+	size_t l1 = len, l2;
+
+	l2 = strlen(s2);
+	if (!l2)
+		return (char *)s1;
+	while (l1 >= l2) {
+		l1--;
+		if (!memcmp(s1, s2, l2))
+			return (char *)s1;
+		s1++;
+	}
+	return NULL;
+}
+EXPORT_SYMBOL(strnstr);
+#endif
+
 #ifndef __HAVE_ARCH_MEMCHR
 /**
  * memchr - Find a character in an area of memory.
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index d4996cf46eb..af4aaa6c36f 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -681,24 +681,55 @@ static char *mac_address_string(char *buf, char *end, u8 *addr,
 	char mac_addr[sizeof("xx:xx:xx:xx:xx:xx")];
 	char *p = mac_addr;
 	int i;
+	char separator;
+
+	if (fmt[1] == 'F') {		/* FDDI canonical format */
+		separator = '-';
+	} else {
+		separator = ':';
+	}
 
 	for (i = 0; i < 6; i++) {
 		p = pack_hex_byte(p, addr[i]);
 		if (fmt[0] == 'M' && i != 5)
-			*p++ = ':';
+			*p++ = separator;
 	}
 	*p = '\0';
 
 	return string(buf, end, mac_addr, spec);
 }
 
-static char *ip4_string(char *p, const u8 *addr, bool leading_zeros)
+static char *ip4_string(char *p, const u8 *addr, const char *fmt)
 {
 	int i;
-
+	bool leading_zeros = (fmt[0] == 'i');
+	int index;
+	int step;
+
+	switch (fmt[2]) {
+	case 'h':
+#ifdef __BIG_ENDIAN
+		index = 0;
+		step = 1;
+#else
+		index = 3;
+		step = -1;
+#endif
+		break;
+	case 'l':
+		index = 3;
+		step = -1;
+		break;
+	case 'n':
+	case 'b':
+	default:
+		index = 0;
+		step = 1;
+		break;
+	}
 	for (i = 0; i < 4; i++) {
 		char temp[3];	/* hold each IP quad in reverse order */
-		int digits = put_dec_trunc(temp, addr[i]) - temp;
+		int digits = put_dec_trunc(temp, addr[index]) - temp;
 		if (leading_zeros) {
 			if (digits < 3)
 				*p++ = '0';
@@ -710,6 +741,7 @@ static char *ip4_string(char *p, const u8 *addr, bool leading_zeros)
 			*p++ = temp[digits];
 		if (i < 3)
 			*p++ = '.';
+		index += step;
 	}
 	*p = '\0';
 
@@ -789,7 +821,7 @@ static char *ip6_compressed_string(char *p, const char *addr)
 	if (useIPv4) {
 		if (needcolon)
 			*p++ = ':';
-		p = ip4_string(p, &in6.s6_addr[12], false);
+		p = ip4_string(p, &in6.s6_addr[12], "I4");
 	}
 	*p = '\0';
 
@@ -829,7 +861,7 @@ static char *ip4_addr_string(char *buf, char *end, const u8 *addr,
 {
 	char ip4_addr[sizeof("255.255.255.255")];
 
-	ip4_string(ip4_addr, addr, fmt[0] == 'i');
+	ip4_string(ip4_addr, addr, fmt);
 
 	return string(buf, end, ip4_addr, spec);
 }
@@ -896,14 +928,17 @@ static char *uuid_string(char *buf, char *end, const u8 *addr,
  * - 'M' For a 6-byte MAC address, it prints the address in the
  *       usual colon-separated hex notation
  * - 'm' For a 6-byte MAC address, it prints the hex address without colons
+ * - 'MF' For a 6-byte MAC FDDI address, it prints the address
+ *       with a dash-separated hex notation
  * - 'I' [46] for IPv4/IPv6 addresses printed in the usual way
  *       IPv4 uses dot-separated decimal without leading 0's (1.2.3.4)
  *       IPv6 uses colon separated network-order 16 bit hex with leading 0's
  * - 'i' [46] for 'raw' IPv4/IPv6 addresses
  *       IPv6 omits the colons (01020304...0f)
  *       IPv4 uses dot-separated decimal with leading 0's (010.123.045.006)
+ * - '[Ii]4[hnbl]' IPv4 addresses in host, network, big or little endian order
  * - 'I6c' for IPv6 addresses printed as specified by
- *       http://www.ietf.org/id/draft-kawamura-ipv6-text-representation-03.txt
+ *       http://tools.ietf.org/html/draft-ietf-6man-text-addr-representation-00
  * - 'U' For a 16 byte UUID/GUID, it prints the UUID/GUID in the form
  *       "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
  *       Options for %pU are:
@@ -939,6 +974,7 @@ static char *pointer(const char *fmt, char *buf, char *end, void *ptr,
 		return resource_string(buf, end, ptr, spec, fmt);
 	case 'M':			/* Colon separated: 00:01:02:03:04:05 */
 	case 'm':			/* Contiguous: 000102030405 */
+					/* [mM]F (FDDI, bit reversed) */
 		return mac_address_string(buf, end, ptr, spec, fmt);
 	case 'I':			/* Formatted IP supported
 					 * 4:	1.2.3.4
@@ -1188,7 +1224,7 @@ qualifier:
  * %pI6 print an IPv6 address with colons
  * %pi6 print an IPv6 address without colons
  * %pI6c print an IPv6 address as specified by
- *   http://www.ietf.org/id/draft-kawamura-ipv6-text-representation-03.txt
+ *   http://tools.ietf.org/html/draft-ietf-6man-text-addr-representation-00
  * %pU[bBlL] print a UUID/GUID in big or little endian using lower or upper
  *   case.
  * %n is ignored
diff --git a/lib/zlib_inflate/inffast.c b/lib/zlib_inflate/inffast.c
index 8550b0c05d0..215447c5526 100644
--- a/lib/zlib_inflate/inffast.c
+++ b/lib/zlib_inflate/inffast.c
@@ -8,6 +8,21 @@
 #include "inflate.h"
 #include "inffast.h"
 
+/* Only do the unaligned "Faster" variant when
+ * CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS is set
+ *
+ * On powerpc, it won't be as we don't include autoconf.h
+ * automatically for the boot wrapper, which is intended as
+ * we run in an environment where we may not be able to deal
+ * with (even rare) alignment faults. In addition, we do not
+ * define __KERNEL__ for arch/powerpc/boot unlike x86
+ */
+
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+#include <asm/unaligned.h>
+#include <asm/byteorder.h>
+#endif
+
 #ifndef ASMINF
 
 /* Allow machine dependent optimization for post-increment or pre-increment.
@@ -24,9 +39,11 @@
 #ifdef POSTINC
 #  define OFF 0
 #  define PUP(a) *(a)++
+#  define UP_UNALIGNED(a) get_unaligned((a)++)
 #else
 #  define OFF 1
 #  define PUP(a) *++(a)
+#  define UP_UNALIGNED(a) get_unaligned(++(a))
 #endif
 
 /*
@@ -239,18 +256,62 @@ void inflate_fast(z_streamp strm, unsigned start)
                     }
                 }
                 else {
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+		    unsigned short *sout;
+		    unsigned long loops;
+
+                    from = out - dist;          /* copy direct from output */
+		    /* minimum length is three */
+		    /* Align out addr */
+		    if (!((long)(out - 1 + OFF) & 1)) {
+			PUP(out) = PUP(from);
+			len--;
+		    }
+		    sout = (unsigned short *)(out - OFF);
+		    if (dist > 2) {
+			unsigned short *sfrom;
+
+			sfrom = (unsigned short *)(from - OFF);
+			loops = len >> 1;
+			do
+			    PUP(sout) = UP_UNALIGNED(sfrom);
+			while (--loops);
+			out = (unsigned char *)sout + OFF;
+			from = (unsigned char *)sfrom + OFF;
+		    } else { /* dist == 1 or dist == 2 */
+			unsigned short pat16;
+
+			pat16 = *(sout-2+2*OFF);
+			if (dist == 1)
+#if defined(__BIG_ENDIAN)
+			    pat16 = (pat16 & 0xff) | ((pat16 & 0xff) << 8);
+#elif defined(__LITTLE_ENDIAN)
+			    pat16 = (pat16 & 0xff00) | ((pat16 & 0xff00) >> 8);
+#else
+#error __BIG_ENDIAN nor __LITTLE_ENDIAN is defined
+#endif
+			loops = len >> 1;
+			do
+			    PUP(sout) = pat16;
+			while (--loops);
+			out = (unsigned char *)sout + OFF;
+		    }
+		    if (len & 1)
+			PUP(out) = PUP(from);
+#else /* CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS */
                     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;
+			 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);
+			 PUP(out) = PUP(from);
+			 if (len > 1)
+			     PUP(out) = PUP(from);
                     }
+#endif /* !CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS */
                 }
             }
             else if ((op & 64) == 0) {          /* 2nd level distance code */