ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/libecb/ecb.pod
(Generate patch)

Comparing libecb/ecb.pod (file contents):
Revision 1.29 by root, Fri Jun 10 12:20:14 2011 UTC vs.
Revision 1.39 by root, Sat Dec 10 11:58:38 2011 UTC

… …
101 #else 101 #else
102 return 0; 102 return 0;
103 #endif 103 #endif
104 } 104 }
105 105
106=item ECB_INLINE 106=item ecb_inline
107 107
108This is not actually an attribute, but you use it like one. It expands 108This is not actually an attribute, but you use it like one. It expands
109either to C<static inline> or to just C<static>, if inline isn't 109either to C<static inline> or to just C<static>, if inline isn't
110supported. It should be used to declare functions that should be inlined, 110supported. It should be used to declare functions that should be inlined,
111for code size or speed reasons. 111for code size or speed reasons.
112 112
113Example: inline this function, it surely will reduce codesize. 113Example: inline this function, it surely will reduce codesize.
114 114
115 ECB_INLINE int 115 ecb_inline int
116 negmul (int a, int b) 116 negmul (int a, int b)
117 { 117 {
118 return - (a * b); 118 return - (a * b);
119 } 119 }
120 120
… …
396After processing the node, (part of) the next node might already be in 396After processing the node, (part of) the next node might already be in
397cache. 397cache.
398 398
399=back 399=back
400 400
401=head2 BIT FIDDLING / BITSTUFFS 401=head2 BIT FIDDLING / BIT WIZARDRY
402 402
403=over 4 403=over 4
404 404
405=item bool ecb_big_endian () 405=item bool ecb_big_endian ()
406 406
… …
412 412
413On systems that are neither, their return values are unspecified. 413On systems that are neither, their return values are unspecified.
414 414
415=item int ecb_ctz32 (uint32_t x) 415=item int ecb_ctz32 (uint32_t x)
416 416
417=item int ecb_ctz64 (uint64_t x)
418
417Returns the index of the least significant bit set in C<x> (or 419Returns the index of the least significant bit set in C<x> (or
418equivalently the number of bits set to 0 before the least significant bit 420equivalently the number of bits set to 0 before the least significant bit
419set), starting from 0. If C<x> is 0 the result is undefined. A common use 421set), starting from 0. If C<x> is 0 the result is undefined.
420case is to compute the integer binary logarithm, i.e., C<floor (log2 422
423For smaller types than C<uint32_t> you can safely use C<ecb_ctz32>.
424
421(n))>. For example: 425For example:
422 426
423 ecb_ctz32 (3) = 0 427 ecb_ctz32 (3) = 0
424 ecb_ctz32 (6) = 1 428 ecb_ctz32 (6) = 1
425 429
430=item int ecb_ld32 (uint32_t x)
431
432=item int ecb_ld64 (uint64_t x)
433
434Returns the index of the most significant bit set in C<x>, or the number
435of digits the number requires in binary (so that C<< 2**ld <= x <
4362**(ld+1) >>). If C<x> is 0 the result is undefined. A common use case is
437to compute the integer binary logarithm, i.e. C<floor (log2 (n))>, for
438example to see how many bits a certain number requires to be encoded.
439
440This function is similar to the "count leading zero bits" function, except
441that that one returns how many zero bits are "in front" of the number (in
442the given data type), while C<ecb_ld> returns how many bits the number
443itself requires.
444
445For smaller types than C<uint32_t> you can safely use C<ecb_ld32>.
446
426=item int ecb_popcount32 (uint32_t x) 447=item int ecb_popcount32 (uint32_t x)
427 448
449=item int ecb_popcount64 (uint64_t x)
450
428Returns the number of bits set to 1 in C<x>. For example: 451Returns the number of bits set to 1 in C<x>.
452
453For smaller types than C<uint32_t> you can safely use C<ecb_popcount32>.
454
455For example:
429 456
430 ecb_popcount32 (7) = 3 457 ecb_popcount32 (7) = 3
431 ecb_popcount32 (255) = 8 458 ecb_popcount32 (255) = 8
432 459
460=item uint8_t ecb_bitrev8 (uint8_t x)
461
462=item uint16_t ecb_bitrev16 (uint16_t x)
463
464=item uint32_t ecb_bitrev32 (uint32_t x)
465
466Reverses the bits in x, i.e. the MSB becomes the LSB, MSB-1 becomes LSB+1
467and so on.
468
469Example:
470
471 ecb_bitrev8 (0xa7) = 0xea
472 ecb_bitrev32 (0xffcc4411) = 0x882233ff
473
433=item uint32_t ecb_bswap16 (uint32_t x) 474=item uint32_t ecb_bswap16 (uint32_t x)
434 475
435=item uint32_t ecb_bswap32 (uint32_t x) 476=item uint32_t ecb_bswap32 (uint32_t x)
436 477
478=item uint64_t ecb_bswap64 (uint64_t x)
479
437These two functions return the value of the 16-bit (32-bit) value C<x> 480These functions return the value of the 16-bit (32-bit, 64-bit) value
438after reversing the order of bytes (0x11223344 becomes 0x44332211). 481C<x> after reversing the order of bytes (0x11223344 becomes 0x44332211 in
482C<ecb_bswap32>).
483
484=item uint8_t ecb_rotl8 (uint8_t x, unsigned int count)
485
486=item uint16_t ecb_rotl16 (uint16_t x, unsigned int count)
487
488=item uint32_t ecb_rotl32 (uint32_t x, unsigned int count)
489
490=item uint64_t ecb_rotl64 (uint64_t x, unsigned int count)
491
492=item uint8_t ecb_rotr8 (uint8_t x, unsigned int count)
493
494=item uint16_t ecb_rotr16 (uint16_t x, unsigned int count)
439 495
440=item uint32_t ecb_rotr32 (uint32_t x, unsigned int count) 496=item uint32_t ecb_rotr32 (uint32_t x, unsigned int count)
441 497
442=item uint32_t ecb_rotl32 (uint32_t x, unsigned int count) 498=item uint64_t ecb_rotr64 (uint64_t x, unsigned int count)
443 499
444These two functions return the value of C<x> after rotating all the bits 500These two families of functions return the value of C<x> after rotating
445by C<count> positions to the right or left respectively. 501all the bits by C<count> positions to the right (C<ecb_rotr>) or left
502(C<ecb_rotl>).
446 503
447Current GCC versions understand these functions and usually compile them 504Current GCC versions understand these functions and usually compile them
448to "optimal" code (e.g. a single C<roll> on x86). 505to "optimal" code (e.g. a single C<rol> or a combination of C<shld> on
506x86).
449 507
450=back 508=back
451 509
452=head2 ARITHMETIC 510=head2 ARITHMETIC
453 511
… …
463C<ecb_mod> implements the mathematical modulo operation, which is missing 521C<ecb_mod> implements the mathematical modulo operation, which is missing
464in the language. 522in the language.
465 523
466C<n> must be strictly positive (i.e. C<< >= 1 >>), while C<m> must be 524C<n> must be strictly positive (i.e. C<< >= 1 >>), while C<m> must be
467negatable, that is, both C<m> and C<-m> must be representable in its 525negatable, that is, both C<m> and C<-m> must be representable in its
468type (this typically includes the minimum signed integer value, the same 526type (this typically excludes the minimum signed integer value, the same
469limitation as for C</> and C<%> in C). 527limitation as for C</> and C<%> in C).
470 528
471Current GCC versions compile this into an efficient branchless sequence on 529Current GCC versions compile this into an efficient branchless sequence on
472almost all CPUs. 530almost all CPUs.
473 531
… …
477change direction for negative values: 535change direction for negative values:
478 536
479 for (m = -100; m <= 100; ++m) 537 for (m = -100; m <= 100; ++m)
480 int elem = myarray [ecb_mod (m, ecb_array_length (myarray))]; 538 int elem = myarray [ecb_mod (m, ecb_array_length (myarray))];
481 539
540=item x = ecb_div_rd (val, div)
541
542=item x = ecb_div_ru (val, div)
543
544Returns C<val> divided by C<div> rounded down or up, respectively.
545C<val> and C<div> must have integer types and C<div> must be strictly
546positive. Note that these functions are implemented with macros in C
547and with function templates in C++.
548
482=back 549=back
483 550
484=head2 UTILITY 551=head2 UTILITY
485 552
486=over 4 553=over 4

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines