diff --git a/src/lzfP.h b/src/lzfP.h
index c9eae3f6..c6d2e096 100644
--- a/src/lzfP.h
+++ b/src/lzfP.h
@@ -49,7 +49,7 @@
* the difference between 15 and 14 is very small
* for small blocks (and 14 is usually a bit faster).
* For a low-memory/faster configuration, use HLOG == 13;
- * For best compression, use 15 or 16 (or more, up to 23).
+ * For best compression, use 15 or 16 (or more, up to 22).
*/
#ifndef HLOG
# define HLOG 16
@@ -94,7 +94,7 @@
/*
* Avoid assigning values to errno variable? for some embedding purposes
* (linux kernel for example), this is necessary. NOTE: this breaks
- * the documentation in lzf.h.
+ * the documentation in lzf.h. Avoiding errno has no speed impact.
*/
#ifndef AVOID_ERRNO
# define AVOID_ERRNO 0
@@ -121,16 +121,52 @@
# define CHECK_INPUT 1
#endif
+/*
+ * Whether to store pointers or offsets inside the hash table. On
+ * 64 bit architetcures, pointers take up twice as much space,
+ * and might also be slower. Default is to autodetect.
+ */
+/*#define LZF_USER_OFFSETS autodetect */
+
/*****************************************************************************/
/* nothing should be changed below */
+#ifdef __cplusplus
+# include <cstring>
+# include <climits>
+using namespace std;
+#else
+# include <string.h>
+# include <limits.h>
+#endif
+
+#ifndef LZF_USE_OFFSETS
+# if defined (WIN32)
+# define LZF_USE_OFFSETS defined(_M_X64)
+# else
+# if __cplusplus > 199711L
+# include <cstdint>
+# else
+# include <stdint.h>
+# endif
+# define LZF_USE_OFFSETS (UINTPTR_MAX > 0xffffffffU)
+# endif
+#endif
+
typedef unsigned char u8;
-typedef const u8 *LZF_STATE[1 << (HLOG)];
+#if LZF_USE_OFFSETS
+# define LZF_HSLOT_BIAS ((const u8 *)in_data)
+ typedef unsigned int LZF_HSLOT;
+#else
+# define LZF_HSLOT_BIAS 0
+ typedef const u8 *LZF_HSLOT;
+#endif
+
+typedef LZF_HSLOT LZF_STATE[1 << (HLOG)];
#if !STRICT_ALIGN
/* for unaligned accesses we need a 16 bit datatype. */
-# include <limits.h>
# if USHRT_MAX == 65535
typedef unsigned short u16;
# elif UINT_MAX == 65535
@@ -142,17 +178,7 @@ typedef const u8 *LZF_STATE[1 << (HLOG)];
#endif
#if ULTRA_FAST
-# if defined(VERY_FAST)
-# undef VERY_FAST
-# endif
-#endif
-
-#if INIT_HTAB
-# ifdef __cplusplus
-# include <cstring>
-# else
-# include <string.h>
-# endif
+# undef VERY_FAST
#endif
#endif
diff --git a/src/lzf_c.c b/src/lzf_c.c
index 9e031ad0..e9c69a0b 100644
--- a/src/lzf_c.c
+++ b/src/lzf_c.c
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2000-2008 Marc Alexander Lehmann <schmorp@schmorp.de>
+ * Copyright (c) 2000-2010 Marc Alexander Lehmann <schmorp@schmorp.de>
*
* Redistribution and use in source and binary forms, with or without modifica-
* tion, are permitted provided that the following conditions are met:
@@ -40,8 +40,8 @@
/*
* don't play with this unless you benchmark!
- * decompression is not dependent on the hash function
- * the hashing function might seem strange, just believe me
+ * the data format is not dependent on the hash function.
+ * the hash function might seem strange, just believe me,
* it works ;)
*/
#ifndef FRST
@@ -89,9 +89,9 @@
/*
* compressed format
*
- * 000LLLLL <L+1> ; literal
- * LLLooooo oooooooo ; backref L
- * 111ooooo LLLLLLLL oooooooo ; backref L+7
+ * 000LLLLL <L+1> ; literal, L+1=1..33 octets
+ * LLLooooo oooooooo ; backref L+1=1..7 octets, o+1=1..4096 offset
+ * 111ooooo LLLLLLLL oooooooo ; backref L+8 octets, o+1=1..4096 offset
*
*/
@@ -106,7 +106,6 @@ lzf_compress (const void *const in_data, unsigned int in_len,
#if !LZF_STATE_ARG
LZF_STATE htab;
#endif
- const u8 **hslot;
const u8 *ip = (const u8 *)in_data;
u8 *op = (u8 *)out_data;
const u8 *in_end = ip + in_len;
@@ -133,10 +132,6 @@ lzf_compress (const void *const in_data, unsigned int in_len,
#if INIT_HTAB
memset (htab, 0, sizeof (htab));
-# if 0
- for (hslot = htab; hslot < htab + HSIZE; hslot++)
- *hslot++ = ip;
-# endif
#endif
lit = 0; op++; /* start run */
@@ -144,24 +139,23 @@ lzf_compress (const void *const in_data, unsigned int in_len,
hval = FRST (ip);
while (ip < in_end - 2)
{
+ LZF_HSLOT *hslot;
+
hval = NEXT (hval, ip);
hslot = htab + IDX (hval);
- ref = *hslot; *hslot = ip;
+ ref = *hslot + LZF_HSLOT_BIAS; *hslot = ip - LZF_HSLOT_BIAS;
if (1
#if INIT_HTAB
&& ref < ip /* the next test will actually take care of this, but this is faster */
#endif
&& (off = ip - ref - 1) < MAX_OFF
- && ip + 4 < in_end
&& ref > (u8 *)in_data
-#if STRICT_ALIGN
- && ref[0] == ip[0]
- && ref[1] == ip[1]
&& ref[2] == ip[2]
+#if STRICT_ALIGN
+ && ((ref[1] << 8) | ref[0]) == ((ip[1] << 8) | ip[0])
#else
&& *(u16 *)ref == *(u16 *)ip
- && ref[2] == ip[2]
#endif
)
{
@@ -170,12 +164,13 @@ lzf_compress (const void *const in_data, unsigned int in_len,
unsigned int maxlen = in_end - ip - len;
maxlen = maxlen > MAX_REF ? MAX_REF : maxlen;
+ if (expect_false (op + 3 + 1 >= out_end)) /* first a faster conservative test */
+ if (op - !lit + 3 + 1 >= out_end) /* second the exact but rare test */
+ return 0;
+
op [- lit - 1] = lit - 1; /* stop run */
op -= !lit; /* undo run if length is zero */
- if (expect_false (op + 3 + 1 >= out_end))
- return 0;
-
for (;;)
{
if (expect_true (maxlen > 16))
@@ -222,6 +217,7 @@ lzf_compress (const void *const in_data, unsigned int in_len,
}
*op++ = off;
+
lit = 0; op++; /* start run */
ip += len + 1;
@@ -237,12 +233,12 @@ lzf_compress (const void *const in_data, unsigned int in_len,
hval = FRST (ip);
hval = NEXT (hval, ip);
- htab[IDX (hval)] = ip;
+ htab[IDX (hval)] = ip - LZF_HSLOT_BIAS;
ip++;
# if VERY_FAST && !ULTRA_FAST
hval = NEXT (hval, ip);
- htab[IDX (hval)] = ip;
+ htab[IDX (hval)] = ip - LZF_HSLOT_BIAS;
ip++;
# endif
#else
@@ -251,7 +247,7 @@ lzf_compress (const void *const in_data, unsigned int in_len,
do
{
hval = NEXT (hval, ip);
- htab[IDX (hval)] = ip;
+ htab[IDX (hval)] = ip - LZF_HSLOT_BIAS;
ip++;
}
while (len--);
diff --git a/src/lzf_d.c b/src/lzf_d.c
index 6c723f5e..c32be8e8 100644
--- a/src/lzf_d.c
+++ b/src/lzf_d.c
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2000-2007 Marc Alexander Lehmann <schmorp@schmorp.de>
+ * Copyright (c) 2000-2010 Marc Alexander Lehmann <schmorp@schmorp.de>
*
* Redistribution and use in source and binary forms, with or without modifica-
* tion, are permitted provided that the following conditions are met:
@@ -43,14 +43,14 @@
# define SET_ERRNO(n) errno = (n)
#endif
-/*
+#if USE_REP_MOVSB /* small win on amd, big loss on intel */
#if (__i386 || __amd64) && __GNUC__ >= 3
# define lzf_movsb(dst, src, len) \
asm ("rep movsb" \
: "=D" (dst), "=S" (src), "=c" (len) \
: "0" (dst), "1" (src), "2" (len));
#endif
-*/
+#endif
unsigned int
lzf_decompress (const void *const in_data, unsigned int in_len,
@@ -86,9 +86,17 @@ lzf_decompress (const void *const in_data, unsigned int in_len,
#ifdef lzf_movsb
lzf_movsb (op, ip, ctrl);
#else
- do
- *op++ = *ip++;
- while (--ctrl);
+ switch (ctrl)
+ {
+ case 32: *op++ = *ip++; case 31: *op++ = *ip++; case 30: *op++ = *ip++; case 29: *op++ = *ip++;
+ case 28: *op++ = *ip++; case 27: *op++ = *ip++; case 26: *op++ = *ip++; case 25: *op++ = *ip++;
+ case 24: *op++ = *ip++; case 23: *op++ = *ip++; case 22: *op++ = *ip++; case 21: *op++ = *ip++;
+ case 20: *op++ = *ip++; case 19: *op++ = *ip++; case 18: *op++ = *ip++; case 17: *op++ = *ip++;
+ case 16: *op++ = *ip++; case 15: *op++ = *ip++; case 14: *op++ = *ip++; case 13: *op++ = *ip++;
+ case 12: *op++ = *ip++; case 11: *op++ = *ip++; case 10: *op++ = *ip++; case 9: *op++ = *ip++;
+ case 8: *op++ = *ip++; case 7: *op++ = *ip++; case 6: *op++ = *ip++; case 5: *op++ = *ip++;
+ case 4: *op++ = *ip++; case 3: *op++ = *ip++; case 2: *op++ = *ip++; case 1: *op++ = *ip++;
+ }
#endif
}
else /* back reference */
@@ -134,12 +142,39 @@ lzf_decompress (const void *const in_data, unsigned int in_len,
len += 2;
lzf_movsb (op, ref, len);
#else
- *op++ = *ref++;
- *op++ = *ref++;
+ switch (len)
+ {
+ default:
+ len += 2;
- do
- *op++ = *ref++;
- while (--len);
+ if (op >= ref + len)
+ {
+ /* disjunct areas */
+ memcpy (op, ref, len);
+ op += len;
+ }
+ else
+ {
+ /* overlapping, use octte by octte copying */
+ do
+ *op++ = *ref++;
+ while (--len);
+ }
+
+ break;
+
+ case 9: *op++ = *ref++;
+ case 8: *op++ = *ref++;
+ case 7: *op++ = *ref++;
+ case 6: *op++ = *ref++;
+ case 5: *op++ = *ref++;
+ case 4: *op++ = *ref++;
+ case 3: *op++ = *ref++;
+ case 2: *op++ = *ref++;
+ case 1: *op++ = *ref++;
+ case 0: *op++ = *ref++; /* two octets more */
+ *op++ = *ref++;
+ }
#endif
}
}