Merge pull request #909 from lz4/fasterCount

Improved software byte count

2 files changed, 71 insertions, 45 deletions

diff --git a/lib/README.md b/lib/README.md
index 707d777..3653c81 100644
--- a/lib/README.md
+++ b/lib/README.md
@@ -46,11 +46,11 @@ and `LZ4F_PUBLISH_STATIC_FUNCTIONS`.
 
 #### Build macros
 
-The following build macro can be selected at compilation time :
+The following build macro can be selected to adjust source code behavior at compilation time :
 
-- `LZ4_FAST_DEC_LOOP` : this triggers the optimized decompression loop.
-  This loops works great on x86/x64 cpus, and is automatically enabled on this platform.
-  It's possible to enable or disable it manually, by passing `LZ4_FAST_DEC_LOOP=1` or `0` to the preprocessor.
+- `LZ4_FAST_DEC_LOOP` : this triggers a speed optimized decompression loop, more powerful on modern cpus.
+  This loop works great on x86, x64 and aarch64 cpus, and is automatically enabled for them.
+  It's also possible to enable or disable it manually, by passing `LZ4_FAST_DEC_LOOP=1` or `0` to the preprocessor.
   For example, with `gcc` : `-DLZ4_FAST_DEC_LOOP=1`,
   and with `make` : `CPPFLAGS+=-DLZ4_FAST_DEC_LOOP=1 make lz4`.
 
@@ -66,9 +66,17 @@ The following build macro can be selected at compilation time :
   Should this be a problem, it's generally possible to make the compiler ignore these warnings,
   for example with `-Wno-deprecated-declarations` on `gcc`,
   or `_CRT_SECURE_NO_WARNINGS` for Visual Studio.
-  Another method is to define `LZ4_DISABLE_DEPRECATE_WARNINGS`
+  Another project-specific method is to define `LZ4_DISABLE_DEPRECATE_WARNINGS`
   before including the LZ4 header files.
 
+- `LZ4_FORCE_SW_BITCOUNT` : by default, the compression algorithm tries to determine lengths
+  by using bitcount instructions, generally implemented as fast single instructions in many cpus.
+  In case the target cpus doesn't support it, or compiler intrinsic doesn't work, or feature bad performance,
+  it's possible to use an optimized software path instead.
+  This is achieved by setting this build macros .
+  In most cases, it's not expected to be necessary,
+  but it can be legitimately considered for less common platforms.
+
 
 #### Amalgamation
 
@@ -103,7 +111,7 @@ The compiled executable will require LZ4 DLL which is available at `dll\liblz4.d
 
 #### Miscellaneous
 
-Other files present in the directory are not source code. There are :
+Other files present in the directory are not source code. They are :
 
  - `LICENSE` : contains the BSD license text
  - `Makefile` : `make` script to compile and install lz4 library (static and dynamic)
diff --git a/lib/lz4.c b/lib/lz4.c
index 70324e2..0ca7b21 100644
--- a/lib/lz4.c
+++ b/lib/lz4.c
@@ -88,6 +88,7 @@
  * Define this parameter if your target system or compiler does not support hardware bit count
  */
 #if defined(_MSC_VER) && defined(_WIN32_WCE)   /* Visual Studio for WinCE doesn't support Hardware bit count */
+#  undef  LZ4_FORCE_SW_BITCOUNT  /* avoid double def */
 #  define LZ4_FORCE_SW_BITCOUNT
 #endif
 
@@ -487,75 +488,92 @@ LZ4_memcpy_using_offset(BYTE* dstPtr, const BYTE* srcPtr, BYTE* dstEnd, const si
 **************************************/
 static unsigned LZ4_NbCommonBytes (reg_t val)
 {
+    assert(val != 0);
     if (LZ4_isLittleEndian()) {
-        if (sizeof(val)==8) {
-#       if defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
+        if (sizeof(val) == 8) {
+#       if defined(_MSC_VER) && (_MSC_VER >= 1800) && defined(_M_AMD64) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            /* x64 CPUS without BMI support interpret `TZCNT` as `REP BSF` */
+            return (unsigned)_tzcnt_u64(val) >> 3;
+#       elif defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
             unsigned long r = 0;
-            _BitScanForward64( &r, (U64)val );
-            return (int)(r>>3);
-#       elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            _BitScanForward64(&r, (U64)val);
+            return (unsigned)r >> 3;
+#       elif (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+                            ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+                                        !defined(LZ4_FORCE_SW_BITCOUNT)
             return (unsigned)__builtin_ctzll((U64)val) >> 3;
 #       else
-            static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2,
-                                                     0, 3, 1, 3, 1, 4, 2, 7,
-                                                     0, 2, 3, 6, 1, 5, 3, 5,
-                                                     1, 3, 4, 4, 2, 5, 6, 7,
-                                                     7, 0, 1, 2, 3, 3, 4, 6,
-                                                     2, 6, 5, 5, 3, 4, 5, 6,
-                                                     7, 1, 2, 4, 6, 4, 4, 5,
-                                                     7, 2, 6, 5, 7, 6, 7, 7 };
-            return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+            const U64 m = 0x0101010101010101ULL;
+            val ^= val - 1;
+            return (unsigned)(((U64)((val & (m - 1)) * m)) >> 56);
 #       endif
         } else /* 32 bits */ {
 #       if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
             unsigned long r;
-            _BitScanForward( &r, (U32)val );
-            return (int)(r>>3);
-#       elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            _BitScanForward(&r, (U32)val);
+            return (unsigned)r >> 3;
+#       elif (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+                            ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+                                        !defined(LZ4_FORCE_SW_BITCOUNT)
             return (unsigned)__builtin_ctz((U32)val) >> 3;
 #       else
-            static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0,
-                                                     3, 2, 2, 1, 3, 2, 0, 1,
-                                                     3, 3, 1, 2, 2, 2, 2, 0,
-                                                     3, 1, 2, 0, 1, 0, 1, 1 };
-            return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+            const U32 m = 0x01010101;
+            return (unsigned)((((val - 1) ^ val) & (m - 1)) * m) >> 24;
 #       endif
         }
     } else   /* Big Endian CPU */ {
-        if (sizeof(val)==8) {   /* 64-bits */
-#       if defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
-            unsigned long r = 0;
-            _BitScanReverse64( &r, val );
-            return (unsigned)(r>>3);
-#       elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
+        if (sizeof(val)==8) {
+#       if (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+                            ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+                                        !defined(LZ4_FORCE_SW_BITCOUNT)
             return (unsigned)__builtin_clzll((U64)val) >> 3;
 #       else
+#if 1
+            /* this method is probably faster,
+             * but adds a 128 bytes lookup table */
+            static const unsigned char ctz7_tab[128] = {
+                7, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                5, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                6, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                5, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+                4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+            };
+            U64 const mask = 0x0101010101010101ULL;
+            U64 const t = (((val >> 8) - mask) | val) & mask;
+            return ctz7_tab[(t * 0x0080402010080402ULL) >> 57];
+#else
+            /* this method doesn't consume memory space like the previous one,
+             * but it contains several branches,
+             * that may end up slowing execution */
             static const U32 by32 = sizeof(val)*4;  /* 32 on 64 bits (goal), 16 on 32 bits.
-                Just to avoid some static analyzer complaining about shift by 32 on 32-bits target.
-                Note that this code path is never triggered in 32-bits mode. */
+            Just to avoid some static analyzer complaining about shift by 32 on 32-bits target.
+            Note that this code path is never triggered in 32-bits mode. */
             unsigned r;
             if (!(val>>by32)) { r=4; } else { r=0; val>>=by32; }
             if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
             r += (!val);
             return r;
+#endif
 #       endif
         } else /* 32 bits */ {
-#       if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
-            unsigned long r = 0;
-            _BitScanReverse( &r, (unsigned long)val );
-            return (unsigned)(r>>3);
-#       elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
+#       if (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+                            ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+                                        !defined(LZ4_FORCE_SW_BITCOUNT)
             return (unsigned)__builtin_clz((U32)val) >> 3;
 #       else
-            unsigned r;
-            if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
-            r += (!val);
-            return r;
+            val >>= 8;
+            val = ((((val + 0x00FFFF00) | 0x00FFFFFF) + val) |
+              (val + 0x00FF0000)) >> 24;
+            return (unsigned)val ^ 3;
 #       endif
         }
     }
 }
 
+
 #define STEPSIZE sizeof(reg_t)
 LZ4_FORCE_INLINE
 unsigned LZ4_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* pInLimit)