ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/libecb/ecb.pod
Revision: 1.56
Committed: Mon Jun 9 17:31:57 2014 UTC (10 years ago) by root
Branch: MAIN
Changes since 1.55: +5 -0 lines
Log Message:
*** empty log message ***

File Contents

# Content
1 =head1 LIBECB - e-C-Builtins
2
3 =head2 ABOUT LIBECB
4
5 Libecb is currently a simple header file that doesn't require any
6 configuration to use or include in your project.
7
8 It's part of the e-suite of libraries, other members of which include
9 libev and libeio.
10
11 Its homepage can be found here:
12
13 http://software.schmorp.de/pkg/libecb
14
15 It mainly provides a number of wrappers around GCC built-ins, together
16 with replacement functions for other compilers. In addition to this,
17 it provides a number of other lowlevel C utilities, such as endianness
18 detection, byte swapping or bit rotations.
19
20 Or in other words, things that should be built into any standard C system,
21 but aren't, implemented as efficient as possible with GCC, and still
22 correct with other compilers.
23
24 More might come.
25
26 =head2 ABOUT THE HEADER
27
28 At the moment, all you have to do is copy F<ecb.h> somewhere where your
29 compiler can find it and include it:
30
31 #include <ecb.h>
32
33 The header should work fine for both C and C++ compilation, and gives you
34 all of F<inttypes.h> in addition to the ECB symbols.
35
36 There are currently no object files to link to - future versions might
37 come with an (optional) object code library to link against, to reduce
38 code size or gain access to additional features.
39
40 It also currently includes everything from F<inttypes.h>.
41
42 =head2 ABOUT THIS MANUAL / CONVENTIONS
43
44 This manual mainly describes each (public) function available after
45 including the F<ecb.h> header. The header might define other symbols than
46 these, but these are not part of the public API, and not supported in any
47 way.
48
49 When the manual mentions a "function" then this could be defined either as
50 as inline function, a macro, or an external symbol.
51
52 When functions use a concrete standard type, such as C<int> or
53 C<uint32_t>, then the corresponding function works only with that type. If
54 only a generic name is used (C<expr>, C<cond>, C<value> and so on), then
55 the corresponding function relies on C to implement the correct types, and
56 is usually implemented as a macro. Specifically, a "bool" in this manual
57 refers to any kind of boolean value, not a specific type.
58
59 =head2 TYPES / TYPE SUPPORT
60
61 ecb.h makes sure that the following types are defined (in the expected way):
62
63 int8_t uint8_t int16_t uint16_t
64 int32_t uint32_t int64_t uint64_t
65 intptr_t uintptr_t
66
67 The macro C<ECB_PTRSIZE> is defined to the size of a pointer on this
68 platform (currently C<4> or C<8>) and can be used in preprocessor
69 expressions.
70
71 For C<ptrdiff_t> and C<size_t> use C<stddef.h>.
72
73 =head2 LANGUAGE/COMPILER VERSIONS
74
75 All the following symbols expand to an expression that can be tested in
76 preprocessor instructions as well as treated as a boolean (use C<!!> to
77 ensure it's either C<0> or C<1> if you need that).
78
79 =over 4
80
81 =item ECB_C
82
83 True if the implementation defines the C<__STDC__> macro to a true value,
84 while not claiming to be C++.
85
86 =item ECB_C99
87
88 True if the implementation claims to be compliant to C99 (ISO/IEC
89 9899:1999) or any later version, while not claiming to be C++.
90
91 Note that later versions (ECB_C11) remove core features again (for
92 example, variable length arrays).
93
94 =item ECB_C11
95
96 True if the implementation claims to be compliant to C11 (ISO/IEC
97 9899:2011) or any later version, while not claiming to be C++.
98
99 =item ECB_CPP
100
101 True if the implementation defines the C<__cplusplus__> macro to a true
102 value, which is typically true for C++ compilers.
103
104 =item ECB_CPP11
105
106 True if the implementation claims to be compliant to ISO/IEC 14882:2011
107 (C++11) or any later version.
108
109 =item ECB_GCC_VERSION(major,minor)
110
111 Expands to a true value (suitable for testing in by the preprocessor)
112 if the compiler used is GNU C and the version is the given version, or
113 higher.
114
115 This macro tries to return false on compilers that claim to be GCC
116 compatible but aren't.
117
118 =item ECB_EXTERN_C
119
120 Expands to C<extern "C"> in C++, and a simple C<extern> in C.
121
122 This can be used to declare a single external C function:
123
124 ECB_EXTERN_C int printf (const char *format, ...);
125
126 =item ECB_EXTERN_C_BEG / ECB_EXTERN_C_END
127
128 These two macros can be used to wrap multiple C<extern "C"> definitions -
129 they expand to nothing in C.
130
131 They are most useful in header files:
132
133 ECB_EXTERN_C_BEG
134
135 int mycfun1 (int x);
136 int mycfun2 (int x);
137
138 ECB_EXTERN_C_END
139
140 =item ECB_STDFP
141
142 If this evaluates to a true value (suitable for testing in by the
143 preprocessor), then C<float> and C<double> use IEEE 754 single/binary32
144 and double/binary64 representations internally I<and> the endianness of
145 both types match the endianness of C<uint32_t> and C<uint64_t>.
146
147 This means you can just copy the bits of a C<float> (or C<double>) to an
148 C<uint32_t> (or C<uint64_t>) and get the raw IEEE 754 bit representation
149 without having to think about format or endianness.
150
151 This is true for basically all modern platforms, although F<ecb.h> might
152 not be able to deduce this correctly everywhere and might err on the safe
153 side.
154
155 =item ECB_AMD64, ECB_AMD64_X32
156
157 These two macros are defined to C<1> on the x86_64/amd64 ABI and the X32
158 ABI, respectively, and undefined elsewhere.
159
160 The designers of the new X32 ABI for some inexplicable reason decided to
161 make it look exactly like amd64, even though it's completely incompatible
162 to that ABI, breaking about every piece of software that assumed that
163 C<__x86_64> stands for, well, the x86-64 ABI, making these macros
164 necessary.
165
166 =back
167
168 =head2 GCC ATTRIBUTES
169
170 A major part of libecb deals with GCC attributes. These are additional
171 attributes that you can assign to functions, variables and sometimes even
172 types - much like C<const> or C<volatile> in C.
173
174 While GCC allows declarations to show up in many surprising places,
175 but not in many expected places, the safest way is to put attribute
176 declarations before the whole declaration:
177
178 ecb_const int mysqrt (int a);
179 ecb_unused int i;
180
181 For variables, it is often nicer to put the attribute after the name, and
182 avoid multiple declarations using commas:
183
184 int i ecb_unused;
185
186 =over 4
187
188 =item ecb_attribute ((attrs...))
189
190 A simple wrapper that expands to C<__attribute__((attrs))> on GCC, and to
191 nothing on other compilers, so the effect is that only GCC sees these.
192
193 Example: use the C<deprecated> attribute on a function.
194
195 ecb_attribute((__deprecated__)) void
196 do_not_use_me_anymore (void);
197
198 =item ecb_unused
199
200 Marks a function or a variable as "unused", which simply suppresses a
201 warning by GCC when it detects it as unused. This is useful when you e.g.
202 declare a variable but do not always use it:
203
204 {
205 int var ecb_unused;
206
207 #ifdef SOMECONDITION
208 var = ...;
209 return var;
210 #else
211 return 0;
212 #endif
213 }
214
215 =item ecb_deprecated
216
217 Similar to C<ecb_unused>, but marks a function, variable or type as
218 deprecated. This makes some compilers warn when the type is used.
219
220 =item ecb_inline
221
222 This is not actually an attribute, but you use it like one. It expands
223 either to C<static inline> or to just C<static>, if inline isn't
224 supported. It should be used to declare functions that should be inlined,
225 for code size or speed reasons.
226
227 Example: inline this function, it surely will reduce codesize.
228
229 ecb_inline int
230 negmul (int a, int b)
231 {
232 return - (a * b);
233 }
234
235 =item ecb_noinline
236
237 Prevent a function from being inlined - it might be optimised away, but
238 not inlined into other functions. This is useful if you know your function
239 is rarely called and large enough for inlining not to be helpful.
240
241 =item ecb_noreturn
242
243 Marks a function as "not returning, ever". Some typical functions that
244 don't return are C<exit> or C<abort> (which really works hard to not
245 return), and now you can make your own:
246
247 ecb_noreturn void
248 my_abort (const char *errline)
249 {
250 puts (errline);
251 abort ();
252 }
253
254 In this case, the compiler would probably be smart enough to deduce it on
255 its own, so this is mainly useful for declarations.
256
257 =item ecb_restrict
258
259 Expands to the C<restrict> keyword or equivalent on compilers that support
260 them, and to nothing on others. Must be specified on a pointer type or
261 an array index to indicate that the memory doesn't alias with any other
262 restricted pointer in the same scope.
263
264 Example: multiply a vector, and allow the compiler to parallelise the
265 loop, because it knows it doesn't overwrite input values.
266
267 void
268 multiply (float *ecb_restrict src,
269 float *ecb_restrict dst,
270 int len, float factor)
271 {
272 int i;
273
274 for (i = 0; i < len; ++i)
275 dst [i] = src [i] * factor;
276 }
277
278 =item ecb_const
279
280 Declares that the function only depends on the values of its arguments,
281 much like a mathematical function. It specifically does not read or write
282 any memory any arguments might point to, global variables, or call any
283 non-const functions. It also must not have any side effects.
284
285 Such a function can be optimised much more aggressively by the compiler -
286 for example, multiple calls with the same arguments can be optimised into
287 a single call, which wouldn't be possible if the compiler would have to
288 expect any side effects.
289
290 It is best suited for functions in the sense of mathematical functions,
291 such as a function returning the square root of its input argument.
292
293 Not suited would be a function that calculates the hash of some memory
294 area you pass in, prints some messages or looks at a global variable to
295 decide on rounding.
296
297 See C<ecb_pure> for a slightly less restrictive class of functions.
298
299 =item ecb_pure
300
301 Similar to C<ecb_const>, declares a function that has no side
302 effects. Unlike C<ecb_const>, the function is allowed to examine global
303 variables and any other memory areas (such as the ones passed to it via
304 pointers).
305
306 While these functions cannot be optimised as aggressively as C<ecb_const>
307 functions, they can still be optimised away in many occasions, and the
308 compiler has more freedom in moving calls to them around.
309
310 Typical examples for such functions would be C<strlen> or C<memcmp>. A
311 function that calculates the MD5 sum of some input and updates some MD5
312 state passed as argument would I<NOT> be pure, however, as it would modify
313 some memory area that is not the return value.
314
315 =item ecb_hot
316
317 This declares a function as "hot" with regards to the cache - the function
318 is used so often, that it is very beneficial to keep it in the cache if
319 possible.
320
321 The compiler reacts by trying to place hot functions near to each other in
322 memory.
323
324 Whether a function is hot or not often depends on the whole program,
325 and less on the function itself. C<ecb_cold> is likely more useful in
326 practise.
327
328 =item ecb_cold
329
330 The opposite of C<ecb_hot> - declares a function as "cold" with regards to
331 the cache, or in other words, this function is not called often, or not at
332 speed-critical times, and keeping it in the cache might be a waste of said
333 cache.
334
335 In addition to placing cold functions together (or at least away from hot
336 functions), this knowledge can be used in other ways, for example, the
337 function will be optimised for size, as opposed to speed, and codepaths
338 leading to calls to those functions can automatically be marked as if
339 C<ecb_expect_false> had been used to reach them.
340
341 Good examples for such functions would be error reporting functions, or
342 functions only called in exceptional or rare cases.
343
344 =item ecb_artificial
345
346 Declares the function as "artificial", in this case meaning that this
347 function is not really meant to be a function, but more like an accessor
348 - many methods in C++ classes are mere accessor functions, and having a
349 crash reported in such a method, or single-stepping through them, is not
350 usually so helpful, especially when it's inlined to just a few instructions.
351
352 Marking them as artificial will instruct the debugger about just this,
353 leading to happier debugging and thus happier lives.
354
355 Example: in some kind of smart-pointer class, mark the pointer accessor as
356 artificial, so that the whole class acts more like a pointer and less like
357 some C++ abstraction monster.
358
359 template<typename T>
360 struct my_smart_ptr
361 {
362 T *value;
363
364 ecb_artificial
365 operator T *()
366 {
367 return value;
368 }
369 };
370
371 =back
372
373 =head2 OPTIMISATION HINTS
374
375 =over 4
376
377 =item bool ecb_is_constant(expr)
378
379 Returns true iff the expression can be deduced to be a compile-time
380 constant, and false otherwise.
381
382 For example, when you have a C<rndm16> function that returns a 16 bit
383 random number, and you have a function that maps this to a range from
384 0..n-1, then you could use this inline function in a header file:
385
386 ecb_inline uint32_t
387 rndm (uint32_t n)
388 {
389 return (n * (uint32_t)rndm16 ()) >> 16;
390 }
391
392 However, for powers of two, you could use a normal mask, but that is only
393 worth it if, at compile time, you can detect this case. This is the case
394 when the passed number is a constant and also a power of two (C<n & (n -
395 1) == 0>):
396
397 ecb_inline uint32_t
398 rndm (uint32_t n)
399 {
400 return is_constant (n) && !(n & (n - 1))
401 ? rndm16 () & (num - 1)
402 : (n * (uint32_t)rndm16 ()) >> 16;
403 }
404
405 =item bool ecb_expect (expr, value)
406
407 Evaluates C<expr> and returns it. In addition, it tells the compiler that
408 the C<expr> evaluates to C<value> a lot, which can be used for static
409 branch optimisations.
410
411 Usually, you want to use the more intuitive C<ecb_expect_true> and
412 C<ecb_expect_false> functions instead.
413
414 =item bool ecb_expect_true (cond)
415
416 =item bool ecb_expect_false (cond)
417
418 These two functions expect a expression that is true or false and return
419 C<1> or C<0>, respectively, so when used in the condition of an C<if> or
420 other conditional statement, it will not change the program:
421
422 /* these two do the same thing */
423 if (some_condition) ...;
424 if (ecb_expect_true (some_condition)) ...;
425
426 However, by using C<ecb_expect_true>, you tell the compiler that the
427 condition is likely to be true (and for C<ecb_expect_false>, that it is
428 unlikely to be true).
429
430 For example, when you check for a null pointer and expect this to be a
431 rare, exceptional, case, then use C<ecb_expect_false>:
432
433 void my_free (void *ptr)
434 {
435 if (ecb_expect_false (ptr == 0))
436 return;
437 }
438
439 Consequent use of these functions to mark away exceptional cases or to
440 tell the compiler what the hot path through a function is can increase
441 performance considerably.
442
443 You might know these functions under the name C<likely> and C<unlikely>
444 - while these are common aliases, we find that the expect name is easier
445 to understand when quickly skimming code. If you wish, you can use
446 C<ecb_likely> instead of C<ecb_expect_true> and C<ecb_unlikely> instead of
447 C<ecb_expect_false> - these are simply aliases.
448
449 A very good example is in a function that reserves more space for some
450 memory block (for example, inside an implementation of a string stream) -
451 each time something is added, you have to check for a buffer overrun, but
452 you expect that most checks will turn out to be false:
453
454 /* make sure we have "size" extra room in our buffer */
455 ecb_inline void
456 reserve (int size)
457 {
458 if (ecb_expect_false (current + size > end))
459 real_reserve_method (size); /* presumably noinline */
460 }
461
462 =item bool ecb_assume (cond)
463
464 Try to tell the compiler that some condition is true, even if it's not
465 obvious.
466
467 This can be used to teach the compiler about invariants or other
468 conditions that might improve code generation, but which are impossible to
469 deduce form the code itself.
470
471 For example, the example reservation function from the C<ecb_expect_false>
472 description could be written thus (only C<ecb_assume> was added):
473
474 ecb_inline void
475 reserve (int size)
476 {
477 if (ecb_expect_false (current + size > end))
478 real_reserve_method (size); /* presumably noinline */
479
480 ecb_assume (current + size <= end);
481 }
482
483 If you then call this function twice, like this:
484
485 reserve (10);
486 reserve (1);
487
488 Then the compiler I<might> be able to optimise out the second call
489 completely, as it knows that C<< current + 1 > end >> is false and the
490 call will never be executed.
491
492 =item bool ecb_unreachable ()
493
494 This function does nothing itself, except tell the compiler that it will
495 never be executed. Apart from suppressing a warning in some cases, this
496 function can be used to implement C<ecb_assume> or similar functions.
497
498 =item bool ecb_prefetch (addr, rw, locality)
499
500 Tells the compiler to try to prefetch memory at the given C<addr>ess
501 for either reading (C<rw> = 0) or writing (C<rw> = 1). A C<locality> of
502 C<0> means that there will only be one access later, C<3> means that
503 the data will likely be accessed very often, and values in between mean
504 something... in between. The memory pointed to by the address does not
505 need to be accessible (it could be a null pointer for example), but C<rw>
506 and C<locality> must be compile-time constants.
507
508 An obvious way to use this is to prefetch some data far away, in a big
509 array you loop over. This prefetches memory some 128 array elements later,
510 in the hope that it will be ready when the CPU arrives at that location.
511
512 int sum = 0;
513
514 for (i = 0; i < N; ++i)
515 {
516 sum += arr [i]
517 ecb_prefetch (arr + i + 128, 0, 0);
518 }
519
520 It's hard to predict how far to prefetch, and most CPUs that can prefetch
521 are often good enough to predict this kind of behaviour themselves. It
522 gets more interesting with linked lists, especially when you do some fair
523 processing on each list element:
524
525 for (node *n = start; n; n = n->next)
526 {
527 ecb_prefetch (n->next, 0, 0);
528 ... do medium amount of work with *n
529 }
530
531 After processing the node, (part of) the next node might already be in
532 cache.
533
534 =back
535
536 =head2 BIT FIDDLING / BIT WIZARDRY
537
538 =over 4
539
540 =item bool ecb_big_endian ()
541
542 =item bool ecb_little_endian ()
543
544 These two functions return true if the byte order is big endian
545 (most-significant byte first) or little endian (least-significant byte
546 first) respectively.
547
548 On systems that are neither, their return values are unspecified.
549
550 =item int ecb_ctz32 (uint32_t x)
551
552 =item int ecb_ctz64 (uint64_t x)
553
554 Returns the index of the least significant bit set in C<x> (or
555 equivalently the number of bits set to 0 before the least significant bit
556 set), starting from 0. If C<x> is 0 the result is undefined.
557
558 For smaller types than C<uint32_t> you can safely use C<ecb_ctz32>.
559
560 For example:
561
562 ecb_ctz32 (3) = 0
563 ecb_ctz32 (6) = 1
564
565 =item bool ecb_is_pot32 (uint32_t x)
566
567 =item bool ecb_is_pot64 (uint32_t x)
568
569 Return true iff C<x> is a power of two or C<x == 0>.
570
571 For smaller types then C<uint32_t> you can safely use C<ecb_is_pot32>.
572
573 =item int ecb_ld32 (uint32_t x)
574
575 =item int ecb_ld64 (uint64_t x)
576
577 Returns the index of the most significant bit set in C<x>, or the number
578 of digits the number requires in binary (so that C<< 2**ld <= x <
579 2**(ld+1) >>). If C<x> is 0 the result is undefined. A common use case is
580 to compute the integer binary logarithm, i.e. C<floor (log2 (n))>, for
581 example to see how many bits a certain number requires to be encoded.
582
583 This function is similar to the "count leading zero bits" function, except
584 that that one returns how many zero bits are "in front" of the number (in
585 the given data type), while C<ecb_ld> returns how many bits the number
586 itself requires.
587
588 For smaller types than C<uint32_t> you can safely use C<ecb_ld32>.
589
590 =item int ecb_popcount32 (uint32_t x)
591
592 =item int ecb_popcount64 (uint64_t x)
593
594 Returns the number of bits set to 1 in C<x>.
595
596 For smaller types than C<uint32_t> you can safely use C<ecb_popcount32>.
597
598 For example:
599
600 ecb_popcount32 (7) = 3
601 ecb_popcount32 (255) = 8
602
603 =item uint8_t ecb_bitrev8 (uint8_t x)
604
605 =item uint16_t ecb_bitrev16 (uint16_t x)
606
607 =item uint32_t ecb_bitrev32 (uint32_t x)
608
609 Reverses the bits in x, i.e. the MSB becomes the LSB, MSB-1 becomes LSB+1
610 and so on.
611
612 Example:
613
614 ecb_bitrev8 (0xa7) = 0xea
615 ecb_bitrev32 (0xffcc4411) = 0x882233ff
616
617 =item uint32_t ecb_bswap16 (uint32_t x)
618
619 =item uint32_t ecb_bswap32 (uint32_t x)
620
621 =item uint64_t ecb_bswap64 (uint64_t x)
622
623 These functions return the value of the 16-bit (32-bit, 64-bit) value
624 C<x> after reversing the order of bytes (0x11223344 becomes 0x44332211 in
625 C<ecb_bswap32>).
626
627 =item uint8_t ecb_rotl8 (uint8_t x, unsigned int count)
628
629 =item uint16_t ecb_rotl16 (uint16_t x, unsigned int count)
630
631 =item uint32_t ecb_rotl32 (uint32_t x, unsigned int count)
632
633 =item uint64_t ecb_rotl64 (uint64_t x, unsigned int count)
634
635 =item uint8_t ecb_rotr8 (uint8_t x, unsigned int count)
636
637 =item uint16_t ecb_rotr16 (uint16_t x, unsigned int count)
638
639 =item uint32_t ecb_rotr32 (uint32_t x, unsigned int count)
640
641 =item uint64_t ecb_rotr64 (uint64_t x, unsigned int count)
642
643 These two families of functions return the value of C<x> after rotating
644 all the bits by C<count> positions to the right (C<ecb_rotr>) or left
645 (C<ecb_rotl>).
646
647 Current GCC versions understand these functions and usually compile them
648 to "optimal" code (e.g. a single C<rol> or a combination of C<shld> on
649 x86).
650
651 =back
652
653 =head2 FLOATING POINT FIDDLING
654
655 =over 4
656
657 =item uint32_t ecb_float_to_binary32 (float x) [-UECB_NO_LIBM]
658
659 =item uint64_t ecb_double_to_binary64 (double x) [-UECB_NO_LIBM]
660
661 These functions each take an argument in the native C<float> or C<double>
662 type and return the IEEE 754 bit representation of it.
663
664 The bit representation is just as IEEE 754 defines it, i.e. the sign bit
665 will be the most significant bit, followed by exponent and mantissa.
666
667 This function should work even when the native floating point format isn't
668 IEEE compliant, of course at a speed and code size penalty, and of course
669 also within reasonable limits (it tries to convert NaNs, infinities and
670 denormals, but will likely convert negative zero to positive zero).
671
672 On all modern platforms (where C<ECB_STDFP> is true), the compiler should
673 be able to optimise away this function completely.
674
675 These functions can be helpful when serialising floats to the network - you
676 can serialise the return value like a normal uint32_t/uint64_t.
677
678 Another use for these functions is to manipulate floating point values
679 directly.
680
681 Silly example: toggle the sign bit of a float.
682
683 /* On gcc-4.7 on amd64, */
684 /* this results in a single add instruction to toggle the bit, and 4 extra */
685 /* instructions to move the float value to an integer register and back. */
686
687 x = ecb_binary32_to_float (ecb_float_to_binary32 (x) ^ 0x80000000U)
688
689 =item float ecb_binary32_to_float (uint32_t x) [-UECB_NO_LIBM]
690
691 =item double ecb_binary32_to_double (uint64_t x) [-UECB_NO_LIBM]
692
693 The reverse operation of the previos function - takes the bit representation
694 of an IEEE binary32 or binary64 number and converts it to the native C<float>
695 or C<double> format.
696
697 This function should work even when the native floating point format isn't
698 IEEE compliant, of course at a speed and code size penalty, and of course
699 also within reasonable limits (it tries to convert normals and denormals,
700 and might be lucky for infinities, and with extraordinary luck, also for
701 negative zero).
702
703 On all modern platforms (where C<ECB_STDFP> is true), the compiler should
704 be able to optimise away this function completely.
705
706 =back
707
708 =head2 ARITHMETIC
709
710 =over 4
711
712 =item x = ecb_mod (m, n)
713
714 Returns C<m> modulo C<n>, which is the same as the positive remainder
715 of the division operation between C<m> and C<n>, using floored
716 division. Unlike the C remainder operator C<%>, this function ensures that
717 the return value is always positive and that the two numbers I<m> and
718 I<m' = m + i * n> result in the same value modulo I<n> - in other words,
719 C<ecb_mod> implements the mathematical modulo operation, which is missing
720 in the language.
721
722 C<n> must be strictly positive (i.e. C<< >= 1 >>), while C<m> must be
723 negatable, that is, both C<m> and C<-m> must be representable in its
724 type (this typically excludes the minimum signed integer value, the same
725 limitation as for C</> and C<%> in C).
726
727 Current GCC versions compile this into an efficient branchless sequence on
728 almost all CPUs.
729
730 For example, when you want to rotate forward through the members of an
731 array for increasing C<m> (which might be negative), then you should use
732 C<ecb_mod>, as the C<%> operator might give either negative results, or
733 change direction for negative values:
734
735 for (m = -100; m <= 100; ++m)
736 int elem = myarray [ecb_mod (m, ecb_array_length (myarray))];
737
738 =item x = ecb_div_rd (val, div)
739
740 =item x = ecb_div_ru (val, div)
741
742 Returns C<val> divided by C<div> rounded down or up, respectively.
743 C<val> and C<div> must have integer types and C<div> must be strictly
744 positive. Note that these functions are implemented with macros in C
745 and with function templates in C++.
746
747 =back
748
749 =head2 UTILITY
750
751 =over 4
752
753 =item element_count = ecb_array_length (name)
754
755 Returns the number of elements in the array C<name>. For example:
756
757 int primes[] = { 2, 3, 5, 7, 11 };
758 int sum = 0;
759
760 for (i = 0; i < ecb_array_length (primes); i++)
761 sum += primes [i];
762
763 =back
764
765 =head2 SYMBOLS GOVERNING COMPILATION OF ECB.H ITSELF
766
767 These symbols need to be defined before including F<ecb.h> the first time.
768
769 =over 4
770
771 =item ECB_NO_THREADS
772
773 If F<ecb.h> is never used from multiple threads, then this symbol can
774 be defined, in which case memory fences (and similar constructs) are
775 completely removed, leading to more efficient code and fewer dependencies.
776
777 Setting this symbol to a true value implies C<ECB_NO_SMP>.
778
779 =item ECB_NO_SMP
780
781 The weaker version of C<ECB_NO_THREADS> - if F<ecb.h> is used from
782 multiple threads, but never concurrently (e.g. if the system the program
783 runs on has only a single CPU with a single core, no hyperthreading and so
784 on), then this symbol can be defined, leading to more efficient code and
785 fewer dependencies.
786
787 =item ECB_NO_LIBM
788
789 When defined to C<1>, do not export any functions that might introduce
790 dependencies on the math library (usually called F<-lm>) - these are
791 marked with [-UECB_NO_LIBM].
792
793 =back
794
795