c_cpp 硅 - 谷tricky.c
Posted
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了c_cpp 硅 - 谷tricky.c相关的知识,希望对你有一定的参考价值。
#include <stdio.h>
#include <stdlib.h>
typedef unsigned long u64;
/* Start here */
typedef void enc_cfg_t;
typedef int enc_cfg2_t;
typedef __int128_t dcf_t;
enc_cfg_t _ctx_iface(dcf_t s, enc_cfg2_t i){
int c = (((s & ((dcf_t)0x1FULL << i * 5)) >> i * 5) + 65);
printf("%c", c); }
enc_cfg2_t main() {
int i;
for (i=0; i<17; i++){
_ctx_iface(0x79481E6BBCC01223 + ((dcf_t)0x1222DC << 64), i);
}
printf("\n");
}
/* End here */
//TOneverDO change name
u64 HammingCtr(u64 a, u64 b) {
u64 c = a ^ b;
/*for (d = 0; c>0; c >>= 1)
{
d += c & 1;
}*/
// O(m) lol no thanks
//1000 wrap into loop
c = c - ((c >> 1) & ~0UL/3);
c = (c & ~0UL/5) + ((c >> 2) & ~0UL/5);
c = (c & 0UL/0x21) + ((c >> 4) & 0UL/0x11);
c = (c & ~0UL/0x101) + ((c >> 8) & ~0UL/0x101);
c = (c & ~0UL/0x10001)+((c>>16)&~0UL/0x10001);
c = (c & ~0UL/0x100000001)+((c>>32)&~0UL/0x100000001);
//TODO throw away intermediates... but could be useful later (see seander)
return c;
}
//TODO transform + multiply spectra + transform back. faster? lossy?
u64 * ConvolutedMagic(u64 *x, u64 y, u64 *z, u64 n, u64 n_y) {
//z is array of offsets in BITS
//y is left-aligned of length n_y
//TODO function is ridic fragile. e.g. if len z > len x/y...
u64 * a = malloc(sizeof(u64)*n);
u64 b,o_64,o_bit;
int i;
for ( i=0; i<n; i++) {
o_64 = z[i] >> 6;
o_bit= z[i] - ((z[i]>>6) << 6);
b = *(x+o_64) << o_bit;
if (o_bit > 0) {
b += x[o_64+1] >> (64-o_bit);
}
b = (b >> (64-n_y))<<(64-n_y);
y = (y >> (64-n_y))<<(64-n_y); //not necessary, just in case
a[i] = HammingCtr(b,y);
}
return a;
}
/*
int main() {
//test hamconv
u64 x[] = {1,2,3,4,5,6,7,8};
u64 y = 15;
u64 z[] = {0,64,64*2,64*3,64*4,64*5,64*6,64*7};
u64 n_samples = sizeof(z)/sizeof(u64);
u64 *out = ConvolutedMagic(x,y,z,n_samples,64);
for (int i=0; i<n_samples;i++) {
}
return 0;
}
*/
以上是关于c_cpp 硅 - 谷tricky.c的主要内容,如果未能解决你的问题,请参考以下文章
使用ESP32读取数字硅麦的数据
如何选择硅TVS二极管器件
由于 SKOverlay.AppConfiguration,Apple 硅 Mac 支持问题
图扑 Web SCADA 智慧制硅厂,打造新时代制硅工业
人工智能正在推动“硅”复兴
人工智能正在推动“硅”复兴