297 lines
8.7 KiB
V
297 lines
8.7 KiB
V
// Copyright (c) 2019 Alexander Medvednikov. All rights reserved.
|
|
// Use of this source code is governed by an MIT license
|
|
// that can be found in the LICENSE file.
|
|
module bits
|
|
|
|
const (
|
|
// See http://supertech.csail.mit.edu/papers/debruijn.pdf
|
|
de_bruijn32 = u32(0x077CB531)
|
|
de_bruijn32tab = [byte(0), 1, 28, 2, 29, 14, 24, 3, 30, 22, 20, 15, 25, 17, 4, 8,
|
|
31, 27, 13, 23, 21, 19, 16, 7, 26, 12, 18, 6, 11, 5, 10, 9,
|
|
]
|
|
de_bruijn64 = (0x03f79d71b4ca8b09)
|
|
de_bruijn64tab = [byte(0), 1, 56, 2, 57, 49, 28, 3, 61, 58, 42, 50, 38, 29, 17, 4,
|
|
62, 47, 59, 36, 45, 43, 51, 22, 53, 39, 33, 30, 24, 18, 12, 5,
|
|
63, 55, 48, 27, 60, 41, 37, 16, 46, 35, 44, 21, 52, 32, 23, 11,
|
|
54, 26, 40, 15, 34, 20, 31, 10, 25, 14, 19, 9, 13, 8, 7, 6,
|
|
]
|
|
)
|
|
|
|
const (
|
|
m0 = 0x5555555555555555 // 01010101 ...
|
|
m1 = 0x3333333333333333 // 00110011 ...
|
|
m2 = 0x0f0f0f0f0f0f0f0f // 00001111 ...
|
|
m3 = 0x00ff00ff00ff00ff // etc.
|
|
m4 = 0x0000ffff0000ffff
|
|
)
|
|
// --- LeadingZeros ---
|
|
// leading_zeros8 returns the number of leading zero bits in x; the result is 8 for x == 0.
|
|
pub fn leading_zeros8(x byte) int {
|
|
return 8 - len8(x)
|
|
}
|
|
|
|
// leading_zeros16 returns the number of leading zero bits in x; the result is 16 for x == 0.
|
|
pub fn leading_zeros16(x u16) int {
|
|
return 16 - len16(x)
|
|
}
|
|
|
|
// leading_zeros32 returns the number of leading zero bits in x; the result is 32 for x == 0.
|
|
pub fn leading_zeros32(x u32) int {
|
|
return 32 - len32(x)
|
|
}
|
|
|
|
// leading_zeros64 returns the number of leading zero bits in x; the result is 64 for x == 0.
|
|
pub fn leading_zeros64(x u64) int {
|
|
return 64 - len64(x)
|
|
}
|
|
|
|
// --- TrailingZeros ---
|
|
// trailing_zeros8 returns the number of trailing zero bits in x; the result is 8 for x == 0.
|
|
pub fn trailing_zeros8(x byte) int {
|
|
return int(ntz8_tab[x])
|
|
}
|
|
|
|
// trailing_zeros16 returns the number of trailing zero bits in x; the result is 16 for x == 0.
|
|
pub fn trailing_zeros16(x u16) int {
|
|
if x == 0 {
|
|
return 16
|
|
}
|
|
// see comment in trailing_zeros64
|
|
return int(de_bruijn32tab[u32(x & -x) * de_bruijn32>>(32 - 5)])
|
|
}
|
|
|
|
// trailing_zeros32 returns the number of trailing zero bits in x; the result is 32 for x == 0.
|
|
pub fn trailing_zeros32(x u32) int {
|
|
if x == 0 {
|
|
return 32
|
|
}
|
|
// see comment in trailing_zeros64
|
|
return int(de_bruijn32tab[(x & -x) * de_bruijn32>>(32 - 5)])
|
|
}
|
|
|
|
// trailing_zeros64 returns the number of trailing zero bits in x; the result is 64 for x == 0.
|
|
pub fn trailing_zeros64(x u64) int {
|
|
if x == 0 {
|
|
return 64
|
|
}
|
|
// If popcount is fast, replace code below with return popcount(^x & (x - 1)).
|
|
//
|
|
// x & -x leaves only the right-most bit set in the word. Let k be the
|
|
// index of that bit. Since only a single bit is set, the value is two
|
|
// to the power of k. Multiplying by a power of two is equivalent to
|
|
// left shifting, in this case by k bits. The de Bruijn (64 bit) constant
|
|
// is such that all six bit, consecutive substrings are distinct.
|
|
// Therefore, if we have a left shifted version of this constant we can
|
|
// find by how many bits it was shifted by looking at which six bit
|
|
// substring ended up at the top of the word.
|
|
// (Knuth, volume 4, section 7.3.1)
|
|
return int(de_bruijn64tab[(x & -x) * de_bruijn64>>(64 - 6)])
|
|
}
|
|
|
|
// --- OnesCount ---
|
|
// ones_count8 returns the number of one bits ("population count") in x.
|
|
pub fn ones_count8(x byte) int {
|
|
return int(pop8_tab[x])
|
|
}
|
|
|
|
// ones_count16 returns the number of one bits ("population count") in x.
|
|
pub fn ones_count16(x u16) int {
|
|
return int(pop8_tab[x>>8] + pop8_tab[x & u16(0xff)])
|
|
}
|
|
|
|
// ones_count32 returns the number of one bits ("population count") in x.
|
|
pub fn ones_count32(x u32) int {
|
|
return int(pop8_tab[x>>24] + pop8_tab[x>>16 & 0xff] + pop8_tab[x>>8 & 0xff] + pop8_tab[x & u32(0xff)])
|
|
}
|
|
|
|
// ones_count64 returns the number of one bits ("population count") in x.
|
|
pub fn ones_count64(x u64) int {
|
|
// Implementation: Parallel summing of adjacent bits.
|
|
// See "Hacker's Delight", Chap. 5: Counting Bits.
|
|
// The following pattern shows the general approach:
|
|
//
|
|
// x = x>>1&(m0&m) + x&(m0&m)
|
|
// x = x>>2&(m1&m) + x&(m1&m)
|
|
// x = x>>4&(m2&m) + x&(m2&m)
|
|
// x = x>>8&(m3&m) + x&(m3&m)
|
|
// x = x>>16&(m4&m) + x&(m4&m)
|
|
// x = x>>32&(m5&m) + x&(m5&m)
|
|
// return int(x)
|
|
//
|
|
// Masking (& operations) can be left away when there's no
|
|
// danger that a field's sum will carry over into the next
|
|
// field: Since the result cannot be > 64, 8 bits is enough
|
|
// and we can ignore the masks for the shifts by 8 and up.
|
|
// Per "Hacker's Delight", the first line can be simplified
|
|
// more, but it saves at best one instruction, so we leave
|
|
// it alone for clarity.
|
|
m := u64(1<<64) - 1
|
|
mut y := (x>>u64(1) & (m0 & m)) + (x & (m0 & m))
|
|
y = (y>>u64(2) & (m1 & m)) + (y & (m1 & m))
|
|
y = ((y>>4) + y) & (m2 & m)
|
|
y += y>>8
|
|
y += y>>16
|
|
y += y>>32
|
|
return int(y) & ((1<<7) - 1)
|
|
}
|
|
|
|
// --- RotateLeft ---
|
|
// rotate_left_8 returns the value of x rotated left by (k mod 8) bits.
|
|
// To rotate x right by k bits, call rotate_left_8(x, -k).
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn rotate_left_8(x byte, k int) byte {
|
|
n := byte(8)
|
|
s := byte(k) & (n - byte(1))
|
|
return ((x<<s) | (x>>(n - s)))
|
|
}
|
|
|
|
// rotate_left_16 returns the value of x rotated left by (k mod 16) bits.
|
|
// To rotate x right by k bits, call rotate_left_16(x, -k).
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn rotate_left_16(x u16, k int) u16 {
|
|
n := u16(16)
|
|
s := u16(k) & (n - u16(1))
|
|
return ((x<<s) | (x>>(n - s)))
|
|
}
|
|
|
|
// rotate_left_32 returns the value of x rotated left by (k mod 32) bits.
|
|
// To rotate x right by k bits, call rotate_left_32(x, -k).
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn rotate_left_32(x u32, k int) u32 {
|
|
n := u32(32)
|
|
s := u32(k) & (n - u32(1))
|
|
return ((x<<s) | (x>>(n - s)))
|
|
}
|
|
|
|
// rotate_left_64 returns the value of x rotated left by (k mod 64) bits.
|
|
// To rotate x right by k bits, call rotate_left_64(x, -k).
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn rotate_left_64(x u64, k int) u64 {
|
|
n := u64(64)
|
|
s := u64(k) & (n - u64(1))
|
|
return ((x<<s) | (x>>(n - s)))
|
|
}
|
|
|
|
// --- Reverse ---
|
|
// reverse8 returns the value of x with its bits in reversed order.
|
|
[inline]
|
|
pub fn reverse8(x byte) byte {
|
|
return rev8_tab[x]
|
|
}
|
|
|
|
// reverse16 returns the value of x with its bits in reversed order.
|
|
[inline]
|
|
pub fn reverse16(x u16) u16 {
|
|
return u16(rev8_tab[x>>8]) | (u16(rev8_tab[x & u16(0xff)])<<8)
|
|
}
|
|
|
|
// reverse32 returns the value of x with its bits in reversed order.
|
|
[inline]
|
|
pub fn reverse32(x u32) u32 {
|
|
m := u64(1<<32) - 1
|
|
mut y := (x>>u32(1) & (m0 & m) | ((x & (m0 & m))<<1))
|
|
y = (y>>u32(2) & (m1 & m) | ((y & (m1 & m))<<u32(2)))
|
|
y = (y>>u32(4) & (m2 & m) | ((y & (m2 & m))<<u32(4)))
|
|
return reverse_bytes32(y)
|
|
}
|
|
|
|
// reverse64 returns the value of x with its bits in reversed order.
|
|
[inline]
|
|
pub fn reverse64(x u64) u64 {
|
|
m := u64(1<<64) - 1
|
|
mut y := (x>>u64(1) & (m0 & m) | ((x & (m0 & m))<<1))
|
|
y = (y>>u64(2) & (m1 & m) | ((y & (m1 & m))<<2))
|
|
y = (y>>u64(4) & (m2 & m) | ((y & (m2 & m))<<4))
|
|
return reverse_bytes64(y)
|
|
}
|
|
|
|
// --- ReverseBytes ---
|
|
// reverse_bytes16 returns the value of x with its bytes in reversed order.
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn reverse_bytes16(x u16) u16 {
|
|
return (x>>8) | (x<<8)
|
|
}
|
|
|
|
// reverse_bytes32 returns the value of x with its bytes in reversed order.
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn reverse_bytes32(x u32) u32 {
|
|
m := u64(1<<32) - 1
|
|
y := (x>>u32(8) & (m3 & m) | ((x & (m3 & m))<<u32(8)))
|
|
return (y>>16) | (y<<16)
|
|
}
|
|
|
|
// reverse_bytes64 returns the value of x with its bytes in reversed order.
|
|
//
|
|
// This function's execution time does not depend on the inputs.
|
|
[inline]
|
|
pub fn reverse_bytes64(x u64) u64 {
|
|
m := u64(1<<64) - 1
|
|
mut y := (x>>u64(8) & (m3 & m) | ((x & (m3 & m))<<u64(8)))
|
|
y = (y>>u64(16) & (m4 & m) | ((y & (m4 & m))<<u64(16)))
|
|
return (y>>32) | (y<<32)
|
|
}
|
|
|
|
// --- Len ---
|
|
// len8 returns the minimum number of bits required to represent x; the result is 0 for x == 0.
|
|
pub fn len8(x byte) int {
|
|
return int(len8_tab[x])
|
|
}
|
|
|
|
// len16 returns the minimum number of bits required to represent x; the result is 0 for x == 0.
|
|
pub fn len16(x u16) int {
|
|
mut y := x
|
|
mut n := 0
|
|
if y >= 1<<8 {
|
|
y >>= 8
|
|
n = 8
|
|
}
|
|
return n + int(len8_tab[y])
|
|
}
|
|
|
|
// len32 returns the minimum number of bits required to represent x; the result is 0 for x == 0.
|
|
pub fn len32(x u32) int {
|
|
mut y := x
|
|
mut n := 0
|
|
if y >= 1<<16 {
|
|
y >>= 16
|
|
n = 16
|
|
}
|
|
if y >= 1<<8 {
|
|
y >>= 8
|
|
n += 8
|
|
}
|
|
return n + int(len8_tab[y])
|
|
}
|
|
|
|
// len64 returns the minimum number of bits required to represent x; the result is 0 for x == 0.
|
|
pub fn len64(x u64) int {
|
|
mut y := x
|
|
mut n := 0
|
|
if y >= u64(1)<<u64(32) {
|
|
y >>= 32
|
|
n = 32
|
|
}
|
|
if y >= u64(1)<<u64(16) {
|
|
y >>= 16
|
|
n += 16
|
|
}
|
|
if y >= u64(1)<<u64(8) {
|
|
y >>= 8
|
|
n += 8
|
|
}
|
|
return n + int(len8_tab[y])
|
|
}
|
|
|