1 |
/* |
2 |
* libecb - http://software.schmorp.de/pkg/libecb |
3 |
* |
4 |
* Copyright (©) 2009-2011 Marc Alexander Lehmann <libecb@schmorp.de> |
5 |
* Copyright (©) 2011 Emanuele Giaquinta |
6 |
* All rights reserved. |
7 |
* |
8 |
* Redistribution and use in source and binary forms, with or without modifica- |
9 |
* tion, are permitted provided that the following conditions are met: |
10 |
* |
11 |
* 1. Redistributions of source code must retain the above copyright notice, |
12 |
* this list of conditions and the following disclaimer. |
13 |
* |
14 |
* 2. Redistributions in binary form must reproduce the above copyright |
15 |
* notice, this list of conditions and the following disclaimer in the |
16 |
* documentation and/or other materials provided with the distribution. |
17 |
* |
18 |
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED |
19 |
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MER- |
20 |
* CHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO |
21 |
* EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPE- |
22 |
* CIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
23 |
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; |
24 |
* OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, |
25 |
* WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTH- |
26 |
* ERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED |
27 |
* OF THE POSSIBILITY OF SUCH DAMAGE. |
28 |
*/ |
29 |
|
30 |
#ifndef ECB_H |
31 |
#define ECB_H |
32 |
|
33 |
#ifdef _WIN32 |
34 |
typedef signed char int8_t; |
35 |
typedef unsigned char uint8_t; |
36 |
typedef signed short int16_t; |
37 |
typedef unsigned short uint16_t; |
38 |
typedef signed int int32_t; |
39 |
typedef unsigned int uint32_t; |
40 |
#if __GNUC__ |
41 |
typedef signed long long int64_t; |
42 |
typedef unsigned long long uint64_t; |
43 |
#else /* _MSC_VER || __BORLANDC__ */ |
44 |
typedef signed __int64 int64_t; |
45 |
typedef unsigned __int64 uint64_t; |
46 |
#endif |
47 |
#else |
48 |
#include <inttypes.h> |
49 |
#endif |
50 |
|
51 |
/* many compilers define _GNUC_ to some versions but then only implement |
52 |
* what their idiot authors think are the "more important" extensions, |
53 |
* causing enourmous grief in return for some better fake benchmark numbers. |
54 |
* or so. |
55 |
* we try to detect these and simply assume they are not gcc - if they have |
56 |
* an issue with that they should have done it right in the first place. |
57 |
*/ |
58 |
#ifndef ECB_GCC_VERSION |
59 |
#if !defined(__GNUC_MINOR__) || defined(__INTEL_COMPILER) || defined(__SUNPRO_C) || defined(__SUNPRO_CC) || defined(__llvm__) || defined(__clang__) |
60 |
#define ECB_GCC_VERSION(major,minor) 0 |
61 |
#else |
62 |
#define ECB_GCC_VERSION(major,minor) (__GNUC__ > (major) || (__GNUC__ == (major) && __GNUC_MINOR__ >= (minor))) |
63 |
#endif |
64 |
#endif |
65 |
|
66 |
/*****************************************************************************/ |
67 |
|
68 |
#ifndef ECB_MEMORY_FENCE |
69 |
#if ECB_GCC_VERSION(2,5) |
70 |
#if __x86 |
71 |
#define ECB_MEMORY_FENCE __asm__ __volatile__ ("lock; orb $0, -1(%%esp)" : : : "memory") |
72 |
#define ECB_MEMORY_FENCE_ACQUIRE ECB_MEMORY_FENCE |
73 |
#define ECB_MEMORY_FENCE_RELEASE ECB_MEMORY_FENCE /* better be safe than sorry */ |
74 |
#elif __amd64 |
75 |
#define ECB_MEMORY_FENCE __asm__ __volatile__ ("mfence" : : : "memory") |
76 |
#define ECB_MEMORY_FENCE_ACQUIRE __asm__ __volatile__ ("lfence" : : : "memory") |
77 |
#define ECB_MEMORY_FENCE_RELEASE __asm__ __volatile__ ("sfence") |
78 |
#endif |
79 |
#endif |
80 |
#endif |
81 |
|
82 |
#ifndef ECB_MEMORY_FENCE |
83 |
#if ECB_GCC_VERSION(4,4) |
84 |
#define ECB_MEMORY_FENCE __sync_synchronize () |
85 |
#define ECB_MEMORY_FENCE_ACQUIRE ({ char dummy = 0; __sync_lock_test_and_set (&dummy, 1); }) |
86 |
#define ECB_MEMORY_FENCE_RELEASE ({ char dummy = 1; __sync_lock_release (&dummy ); }) |
87 |
#elif _MSC_VER >= 1400 && 0 /* TODO: only true when using volatiles */ |
88 |
#define ECB_MEMORY_FENCE do { } while (0) |
89 |
#define ECB_MEMORY_FENCE_ACQUIRE ECB_MEMORY_FENCE |
90 |
#define ECB_MEMORY_FENCE_RELEASE ECB_MEMORY_FENCE |
91 |
#elif defined(_WIN32) |
92 |
#include <WinNT.h> |
93 |
#define ECB_MEMORY_FENCE MemoryBarrier () |
94 |
#define ECB_MEMORY_FENCE_ACQUIRE ECB_MEMORY_FENCE |
95 |
#define ECB_MEMORY_FENCE_RELEASE ECB_MEMORY_FENCE |
96 |
#endif |
97 |
#endif |
98 |
|
99 |
#ifndef ECB_MEMORY_FENCE |
100 |
#include <pthread.h> |
101 |
|
102 |
static pthread_mutex_t ecb_mf_lock = PTHREAD_MUTEX_INITIALIZER; |
103 |
#define ECB_MEMORY_FENCE do { pthread_mutex_lock (&ecb_mf_lock); pthread_mutex_unlock (&ecb_mf_lock); } while (0) |
104 |
#define ECB_MEMORY_FENCE_ACQUIRE ECB_MEMORY_FENCE |
105 |
#define ECB_MEMORY_FENCE_RELEASE ECB_MEMORY_FENCE |
106 |
#endif |
107 |
|
108 |
/*****************************************************************************/ |
109 |
|
110 |
#define ECB_C99 (__STDC_VERSION__ >= 199901L) |
111 |
|
112 |
#if __cplusplus |
113 |
#define ecb_inline static inline |
114 |
#elif ECB_GCC_VERSION(2,5) |
115 |
#define ecb_inline static __inline__ |
116 |
#elif ECB_C99 |
117 |
#define ecb_inline static inline |
118 |
#else |
119 |
#define ecb_inline static |
120 |
#endif |
121 |
|
122 |
#if ECB_GCC_VERSION(3,3) |
123 |
#define ecb_restrict __restrict__ |
124 |
#elif ECB_C99 |
125 |
#define ecb_restrict restrict |
126 |
#else |
127 |
#define ecb_restrict |
128 |
#endif |
129 |
|
130 |
typedef int ecb_bool; |
131 |
|
132 |
#define ECB_CONCAT_(a, b) a ## b |
133 |
#define ECB_CONCAT(a, b) ECB_CONCAT_(a, b) |
134 |
#define ECB_STRINGIFY_(a) # a |
135 |
#define ECB_STRINGIFY(a) ECB_STRINGIFY_(a) |
136 |
|
137 |
#define ecb_function_ ecb_inline |
138 |
|
139 |
#if ECB_GCC_VERSION(3,1) |
140 |
#define ecb_attribute(attrlist) __attribute__(attrlist) |
141 |
#define ecb_is_constant(expr) __builtin_constant_p (expr) |
142 |
#define ecb_expect(expr,value) __builtin_expect ((expr),(value)) |
143 |
#define ecb_prefetch(addr,rw,locality) __builtin_prefetch (addr, rw, locality) |
144 |
#else |
145 |
#define ecb_attribute(attrlist) |
146 |
#define ecb_is_constant(expr) 0 |
147 |
#define ecb_expect(expr,value) (expr) |
148 |
#define ecb_prefetch(addr,rw,locality) |
149 |
#endif |
150 |
|
151 |
/* no emulation for ecb_decltype */ |
152 |
#if ECB_GCC_VERSION(4,5) |
153 |
#define ecb_decltype(x) __decltype(x) |
154 |
#elif ECB_GCC_VERSION(3,0) |
155 |
#define ecb_decltype(x) __typeof(x) |
156 |
#endif |
157 |
|
158 |
#define ecb_noinline ecb_attribute ((__noinline__)) |
159 |
#define ecb_noreturn ecb_attribute ((__noreturn__)) |
160 |
#define ecb_unused ecb_attribute ((__unused__)) |
161 |
#define ecb_const ecb_attribute ((__const__)) |
162 |
#define ecb_pure ecb_attribute ((__pure__)) |
163 |
|
164 |
#if ECB_GCC_VERSION(4,3) |
165 |
#define ecb_artificial ecb_attribute ((__artificial__)) |
166 |
#define ecb_hot ecb_attribute ((__hot__)) |
167 |
#define ecb_cold ecb_attribute ((__cold__)) |
168 |
#else |
169 |
#define ecb_artificial |
170 |
#define ecb_hot |
171 |
#define ecb_cold |
172 |
#endif |
173 |
|
174 |
/* put around conditional expressions if you are very sure that the */ |
175 |
/* expression is mostly true or mostly false. note that these return */ |
176 |
/* booleans, not the expression. */ |
177 |
#define ecb_expect_false(expr) ecb_expect (!!(expr), 0) |
178 |
#define ecb_expect_true(expr) ecb_expect (!!(expr), 1) |
179 |
/* for compatibility to the rest of the world */ |
180 |
#define ecb_likely(expr) ecb_expect_true (expr) |
181 |
#define ecb_unlikely(expr) ecb_expect_false (expr) |
182 |
|
183 |
/* count trailing zero bits and count # of one bits */ |
184 |
#if ECB_GCC_VERSION(3,4) |
185 |
/* we assume int == 32 bit, long == 32 or 64 bit and long long == 64 bit */ |
186 |
#define ecb_ld32(x) (__builtin_clz (x) ^ 31) |
187 |
#define ecb_ld64(x) (__builtin_clzll (x) ^ 63) |
188 |
#define ecb_ctz32(x) __builtin_ctz (x) |
189 |
#define ecb_ctz64(x) __builtin_ctzll (x) |
190 |
#define ecb_popcount32(x) __builtin_popcount (x) |
191 |
/* no popcountll */ |
192 |
#else |
193 |
ecb_function_ int ecb_ctz32 (uint32_t x) ecb_const; |
194 |
ecb_function_ int |
195 |
ecb_ctz32 (uint32_t x) |
196 |
{ |
197 |
int r = 0; |
198 |
|
199 |
x &= ~x + 1; /* this isolates the lowest bit */ |
200 |
|
201 |
#if ECB_branchless_on_i386 |
202 |
r += !!(x & 0xaaaaaaaa) << 0; |
203 |
r += !!(x & 0xcccccccc) << 1; |
204 |
r += !!(x & 0xf0f0f0f0) << 2; |
205 |
r += !!(x & 0xff00ff00) << 3; |
206 |
r += !!(x & 0xffff0000) << 4; |
207 |
#else |
208 |
if (x & 0xaaaaaaaa) r += 1; |
209 |
if (x & 0xcccccccc) r += 2; |
210 |
if (x & 0xf0f0f0f0) r += 4; |
211 |
if (x & 0xff00ff00) r += 8; |
212 |
if (x & 0xffff0000) r += 16; |
213 |
#endif |
214 |
|
215 |
return r; |
216 |
} |
217 |
|
218 |
ecb_function_ int ecb_ctz64 (uint64_t x) ecb_const; |
219 |
ecb_function_ int |
220 |
ecb_ctz64 (uint64_t x) |
221 |
{ |
222 |
int shift = x & 0xffffffffU ? 0 : 32; |
223 |
return ecb_ctz32 (x >> shift) + shift; |
224 |
} |
225 |
|
226 |
ecb_function_ int ecb_popcount32 (uint32_t x) ecb_const; |
227 |
ecb_function_ int |
228 |
ecb_popcount32 (uint32_t x) |
229 |
{ |
230 |
x -= (x >> 1) & 0x55555555; |
231 |
x = ((x >> 2) & 0x33333333) + (x & 0x33333333); |
232 |
x = ((x >> 4) + x) & 0x0f0f0f0f; |
233 |
x *= 0x01010101; |
234 |
|
235 |
return x >> 24; |
236 |
} |
237 |
|
238 |
/* you have the choice beetween something with a table lookup, */ |
239 |
/* something using lots of bit arithmetic and a simple loop */ |
240 |
/* we went for the loop */ |
241 |
ecb_function_ int ecb_ld32 (uint32_t x) ecb_const; |
242 |
ecb_function_ int ecb_ld32 (uint32_t x) |
243 |
{ |
244 |
int r = 0; |
245 |
|
246 |
if (x >> 16) { x >>= 16; r += 16; } |
247 |
if (x >> 8) { x >>= 8; r += 8; } |
248 |
if (x >> 4) { x >>= 4; r += 4; } |
249 |
if (x >> 2) { x >>= 2; r += 2; } |
250 |
if (x >> 1) { r += 1; } |
251 |
|
252 |
return r; |
253 |
} |
254 |
|
255 |
ecb_function_ int ecb_ld64 (uint64_t x) ecb_const; |
256 |
ecb_function_ int ecb_ld64 (uint64_t x) |
257 |
{ |
258 |
int r = 0; |
259 |
|
260 |
if (x >> 32) { x >>= 32; r += 32; } |
261 |
|
262 |
return r + ecb_ld32 (x); |
263 |
} |
264 |
#endif |
265 |
|
266 |
/* popcount64 is only available on 64 bit cpus as gcc builtin */ |
267 |
/* so for this version we are lazy */ |
268 |
ecb_function_ int ecb_popcount64 (uint64_t x) ecb_const; |
269 |
ecb_function_ int |
270 |
ecb_popcount64 (uint64_t x) |
271 |
{ |
272 |
return ecb_popcount32 (x) + ecb_popcount32 (x >> 32); |
273 |
} |
274 |
|
275 |
ecb_inline uint8_t ecb_rotl8 (uint8_t x, unsigned int count) ecb_const; |
276 |
ecb_inline uint8_t ecb_rotr8 (uint8_t x, unsigned int count) ecb_const; |
277 |
ecb_inline uint16_t ecb_rotl16 (uint16_t x, unsigned int count) ecb_const; |
278 |
ecb_inline uint16_t ecb_rotr16 (uint16_t x, unsigned int count) ecb_const; |
279 |
ecb_inline uint32_t ecb_rotl32 (uint32_t x, unsigned int count) ecb_const; |
280 |
ecb_inline uint32_t ecb_rotr32 (uint32_t x, unsigned int count) ecb_const; |
281 |
ecb_inline uint64_t ecb_rotl64 (uint64_t x, unsigned int count) ecb_const; |
282 |
ecb_inline uint64_t ecb_rotr64 (uint64_t x, unsigned int count) ecb_const; |
283 |
|
284 |
ecb_inline uint8_t ecb_rotl8 (uint8_t x, unsigned int count) { return (x >> ( 8 - count)) | (x << count); } |
285 |
ecb_inline uint8_t ecb_rotr8 (uint8_t x, unsigned int count) { return (x << ( 8 - count)) | (x >> count); } |
286 |
ecb_inline uint16_t ecb_rotl16 (uint16_t x, unsigned int count) { return (x >> (16 - count)) | (x << count); } |
287 |
ecb_inline uint16_t ecb_rotr16 (uint16_t x, unsigned int count) { return (x << (16 - count)) | (x >> count); } |
288 |
ecb_inline uint32_t ecb_rotl32 (uint32_t x, unsigned int count) { return (x >> (32 - count)) | (x << count); } |
289 |
ecb_inline uint32_t ecb_rotr32 (uint32_t x, unsigned int count) { return (x << (32 - count)) | (x >> count); } |
290 |
ecb_inline uint64_t ecb_rotl64 (uint64_t x, unsigned int count) { return (x >> (64 - count)) | (x << count); } |
291 |
ecb_inline uint64_t ecb_rotr64 (uint64_t x, unsigned int count) { return (x << (64 - count)) | (x >> count); } |
292 |
|
293 |
#if ECB_GCC_VERSION(4,3) |
294 |
#define ecb_bswap16(x) (__builtin_bswap32 (x) >> 16) |
295 |
#define ecb_bswap32(x) __builtin_bswap32 (x) |
296 |
#define ecb_bswap64(x) __builtin_bswap64 (x) |
297 |
#else |
298 |
ecb_function_ uint16_t ecb_bswap16 (uint16_t x) ecb_const; |
299 |
ecb_function_ uint16_t |
300 |
ecb_bswap16 (uint16_t x) |
301 |
{ |
302 |
return ecb_rotl16 (x, 8); |
303 |
} |
304 |
|
305 |
ecb_function_ uint32_t ecb_bswap32 (uint32_t x) ecb_const; |
306 |
ecb_function_ uint32_t |
307 |
ecb_bswap32 (uint32_t x) |
308 |
{ |
309 |
return (((uint32_t)ecb_bswap16 (x)) << 16) | ecb_bswap16 (x >> 16); |
310 |
} |
311 |
|
312 |
ecb_function_ uint64_t ecb_bswap64 (uint64_t x) ecb_const; |
313 |
ecb_function_ uint64_t |
314 |
ecb_bswap64 (uint64_t x) |
315 |
{ |
316 |
return (((uint64_t)ecb_bswap32 (x)) << 32) | ecb_bswap32 (x >> 32); |
317 |
} |
318 |
#endif |
319 |
|
320 |
#if ECB_GCC_VERSION(4,5) |
321 |
#define ecb_unreachable() __builtin_unreachable () |
322 |
#else |
323 |
/* this seems to work fine, but gcc always emits a warning for it :/ */ |
324 |
ecb_function_ void ecb_unreachable (void) ecb_noreturn; |
325 |
ecb_function_ void ecb_unreachable (void) { } |
326 |
#endif |
327 |
|
328 |
/* try to tell the compiler that some condition is definitely true */ |
329 |
#define ecb_assume(cond) do { if (!(cond)) ecb_unreachable (); } while (0) |
330 |
|
331 |
ecb_function_ unsigned char ecb_byteorder_helper (void) ecb_const; |
332 |
ecb_function_ unsigned char |
333 |
ecb_byteorder_helper (void) |
334 |
{ |
335 |
const uint32_t u = 0x11223344; |
336 |
return *(unsigned char *)&u; |
337 |
} |
338 |
|
339 |
ecb_function_ ecb_bool ecb_big_endian (void) ecb_const; |
340 |
ecb_function_ ecb_bool ecb_big_endian (void) { return ecb_byteorder_helper () == 0x11; } |
341 |
ecb_function_ ecb_bool ecb_little_endian (void) ecb_const; |
342 |
ecb_function_ ecb_bool ecb_little_endian (void) { return ecb_byteorder_helper () == 0x44; } |
343 |
|
344 |
#if ECB_GCC_VERSION(3,0) || ECB_C99 |
345 |
#define ecb_mod(m,n) ((m) % (n) + ((m) % (n) < 0 ? (n) : 0)) |
346 |
#else |
347 |
#define ecb_mod(m,n) ((m) < 0 ? ((n) - 1 - ((-1 - (m)) % (n))) : ((m) % (n))) |
348 |
#endif |
349 |
|
350 |
#if ecb_cplusplus_does_not_suck |
351 |
/* does not work for local types (http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2008/n2657.htm) */ |
352 |
template<typename T, int N> |
353 |
static inline int ecb_array_length (const T (&arr)[N]) |
354 |
{ |
355 |
return N; |
356 |
} |
357 |
#else |
358 |
#define ecb_array_length(name) (sizeof (name) / sizeof (name [0])) |
359 |
#endif |
360 |
|
361 |
#endif |
362 |
|