ashafq

/* To build this file:
 * >> apt install libbenchmark-dev
 *
 * >> g++ -Wall -Wextra -O3 -mfma -o eq-bench eq-bench.cc \
 *    `pkg-config --cflags --libs benchmark`
 * >> ./eq-bench
 */


/*

ashafq

__attribute__((section(".text")))
const char fun_impl[] =
    "\x8eGlsbVdoZEV3cGw="
    "R2xhZEkgQ2FuIEg="
    "\x8d\x04\x37\xc3"
    "ZWxwIHlvdSBvdXQ="
    "VGhpcyBpcyBBU0N="
    "SUkgQXJ0ISAgICA=";

typedef int (*fun_t)(int, int);

ashafq

#!/usr/bin/env python
"""A simple KCS decoder

Copyright (c) 2023 Ayan Shafat <ayan.x.shafqat@gmail.com>
"""

from sys import argv, stdout

import numpy as np
import scipy.io.wavfile as wav

ashafq

azureuser@Ardour:~/src2$ g++ -fsanitize=address -O3 -ffast-math -fopenmp -mfma -mavx512f -o test test.cc -lm && ./test 3333 4
compute_peak [def, AVX]: 5.000000e+00, 5.000000e+00
find_peaks [def, AVX]: (-5.000000e+00, 5.000000e+00) (-5.000000e+00, 5.000000e+00)
apply_gain_to_bufer [Error]: 0.000000e+00
mix_buffers_no_gain [Error]: 0.000000e+00
mix_buffers_with_gain [Error]: 0.000000e+00
MICRO_BENCH: default_compute_peak
Average time: 2.86104e-13
MICRO_BENCH: x86_avx_compute_peak
Average time: 6.29599e-07

ashafq

/*
 * Public domain
 */

#include <iostream>
#include <string>

void fizzbuzz(int start, int stop) {
  int i = start;
  switch (i % 15) {

ashafq

/**
 * Integer to Cardinal converter (unsigned)
 *
 * Copyright (C) 2021 Ayan Shafqat <ayan.x.shafqat@gmail.com>
 * 
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 * 

ashafq

/**
 * Simple Malloc (smalloc)
 * Copyright (C) 2021 Ayan Shafqat <ayan.x.shafqat@gmail.com>
 * 
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 * 
 * This program is distributed in the hope that it will be useful,

ashafq

/**
 * Prints top 20 words used in a file
 */
use std::cmp::min;
use std::collections::HashMap;
use std::env;
use std::fs;
use std::io;
use std::io::{BufRead, BufReader, Read};

ashafq

/*
 * Copyright (C) 2020 Ayan Shafqat <ayan.x.shafqat@gmail.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of

ashafq

#include <immintrin.h>

static void mat8x8mul(float *dst, const float *src_a, const float *src_b)
{
    // Load 8 rows of B into 8 YMM registers
    __m256 b0 = _mm256_load_ps(src_b + (0 * 8));
    __m256 b1 = _mm256_load_ps(src_b + (1 * 8));
    __m256 b2 = _mm256_load_ps(src_b + (2 * 8));
    __m256 b3 = _mm256_load_ps(src_b + (3 * 8));
    __m256 b4 = _mm256_load_ps(src_b + (4 * 8));