Find largest prime number below input value - c

Let me start by saying that I am a beginner. I'm trying to define a function that calculates/identifies the largest prime number below an input value. However, my current approach is flawed.
I've tried implementing a nested for loop. Generating numbers from one below the input down to 1, subsequently running each number through the second loop so as to identify whether or not it is prime. If it is prime (if count == 2) the function is supposed to return the number that was generated by the first loop (n)
I've been permitted to assume that the input will be a positive integer greater than 2.
int prime(int maximum)
{
int i, j, count = 0, n;
for (i = 1; i < maximum; i++) {
n = maximum - i; /* generating number below input value*/
for (j = 1; j <= maximum; j++) {
if (n % j == 0) { /* testing whether or not it is prime */
count++;
}
} if(count == 2) {
break;
}
}
return n;
}
I'd expect an input of 10 to produce an output of 7, an input of 30 to produce an output of 29 and an input of 100 to produce an output of 97.
However, the function is currently generating an output of 1 - consistently.
The code is not generating any error messages
note: This is my first time utilising this platform, my most sincere apologies if the formatting of my question is incorrect

The problem is you are not resetting the count to 0 in the outer loop.
This will solve the problem.
Adding further,
There are some other optimizations that can be done in the code too.
You need not run the inner loop maximum times, only n times will be sufficient as you are checking whether n is prime or not.
You can put the break condition inside the inner for loop for those whose count has crossed 2 to avoid further iterations.
int prime(int maximum)
{
int i, j, count = 0, n;
for (i = 1; i < maximum; i++) {
count = 0; # <----- Reset it to 0.
n = maximum - i; /* generating number below input value*/
for (j = 1; j <= n; j++) {
if (n % j == 0) { /* testing whether or not it is prime */
count++;
if(count > 2) # to avoid further iterations as we know, this number is not prime.
break;
}
} if(count == 2) {
break;
}
}
return n;
}

Related

counting prime numbers infinite loop

I tried counting how many prime there are (except 1 and 0) until there are N number of primes. But somehow my program always ends up looping INFINITELY
int main (){
int n;
printf("Enter size N: ");
scanf("%d", &n);
int i, j, ctr = 0, flag = 0;
for (i = 2; ctr != n; i++){
for (j = 2; j < i; j++){
if (i%j==0){
flag = 1;
break;
}
}
if(flag!=1){
ctr++;
}
}
}
I notice you never reset your flag.
So once a divider is found, you raise your flag to indicate it isn't a prime.
But never set it back to 0, so all number afterward are considered not prime
int i, j;
int ctr = 0; // Prime number counter
int flag;
// For i, from 2 to whatever needed to have N prime numbers
for (i = 2; ctr != n; i++){
flag = 0 // THIS IS THE LINE YOU'RE MISSING
// Look for potential divider
for (j = 2; j < i; j++){
// If i is divided by j, it isn't prime
if (i%j==0){
flag = 1;
break;
}
}
// If no divider found, i is prime
if(flag!=1){
ctr++;
}
}
To add something to your program different than what Portevent mentioned, you dont have to check if i % j != 0 for all j < i to check if i is prime. You just have to check for the range j < sqrt(i).
This is because you stop looping when you get a j value such that i % j == 0. If you were to get to j == sqrt(i) it means for i to not be prime it would have at least a prime factorization of p * q with p and q being greater than sqrt(i), which is nonsense, hence i is prime.
PS: in case you add this optimization to your program, do not hardcode sqrt(i) in the for loop, assign it to some variable then write the variable in the loop to avoid repeated computations.
Who told you that your program loops infinitely? It doesnt.
You just missed resetting the flag like others specified. And also:
you can loop for j < i/2;
Think about it: what is the biggest integer divisor of a prime/non-prime number other than self?
To find it out divide it by the possible minimum integer number:
if we divide it by 1 we get the number itself *
if we divide it by 2 we get the half of the number which is the possible maximum divisor.

Prime Number between given interval

Given below is the code for finding prime numbers between the interval entered by the user.
#include <stdio.h>
int main() {
int n1, n2, i, flag;
scanf("%d%d", &n1, &n2);
for (i = n1; i <= n2; i++) {
flag = prime(i);
if (flag == 1)
printf("\n%d", i);
}
return 0;
}
int prime(int n) {
int j, flag = 1;
for (j = 2; j <= n / 2; j++) {
if (n % j == 0) {
flag = 0;
break;
}
}
return flag;
}
Can anyone explain me how this code deals with odd number, which are not prime (for ex: 15, 21, 25, etc)
int prime(int n) {
int j, flag = 1;
for (j = 2; j <= n / 2; j++) {
if (n % j == 0) {
flag = 0;
break;
}
}
return flag;
}
See in this prime function, when we observe the iteration of for loop if value of n is 15 then it will look like this:
for (j = 2; j <= 15 / 2; j++)
I agree this is true. Because 2<7.
Since the condition is true we will enter inside the for loop:
if(n%j==0){
flag=0;
break;
}
Now, since n=15 and j=2, value of n%j=1, which is obviously not equals to 0; so if loop will not be executed and the prime function will return flag =1; and the main function will print 15 as a prime.
But, after Executing the program the code is showing the correct results: it's not showing 15 as a prime.
So can anyone please help me understand the logic behind this code? (Actually I want to understand how this code is eliminating non-prime odd numbers.)
You checked the execution for j==2, but since there is a for loop for(j=2;j<=n/2;j++). The code will run from j=2 to j=n/2. So, if you consider all the iterations, you will realize that the function is working fine.
The first if statement is false, so for j==2, the program won't go inside the if statement.
The loop will iterate for the next value of j, which is 3. Since 15%3 == 0, the program will execute the statements within the if statement and return that 15 is not a prime number.
for(j=2;j<=n/2;j++){
if(n%j==0){
flag=0;
break;
}
}
In the case of n=15, the loop starts at i=2, the test i<=n/2 is true because 2<=7, then 15%2 is 1, hence the loop proceeds and i is incremented to 3, the loop test is true again because 3<=7 but 15%3 is 0 so flag is set to 0 and returned.
Note these remarks:
the code does not have a recursive function. You merely call a function prime() to check each number in the interval for primality.
prime() should be defined or at least declared before the main() function that calls it.
you can test the return value of prime(i) directly. No need for a flag variable.
for prime numbers, the loop will iterate way too far: you can change the test to j <= n / j to stop at the square root of n.
you can return directly from the loop body.
you should output the newline after the number.
Here is a modified version:
#include <stdio.h>
int isprime(int n) {
int j;
for (j = 2; j <= n / j; j++) {
if (n % j == 0)
return 0;
}
return 1;
}
int main() {
int n1, n2, i;
if (scanf("%d%d", &n1, &n2) != 2)
return 1;
for (i = n1; i <= n2; i++) {
if (isprime(i))
printf("%d\n", i);
}
return 0;
}
Can anyone explain me how this code deals with odd number, which are not prime (for ex: 15, 21, 25, etc)
int prime(int n) {
int j, flag = 1;
for (j = 2; j <= n / 2; j++) {
if (n % j == 0) {
flag = 0;
break;
}
}
return flag;
}
Well this function doesn't need to handle specially nonprime numbers, based on the fact that if we can divide the number n by something (be prime or not), the number will be compose. What it does it to get out of the loop (with flag changed into 0) as soon as it finds a number j that divides n.
There's an extra optimization, that can save you a lot of time, that consists on calculating numbers until the integer rounded down square root of n as, if you can divide the number by a number that is greater than the square root, for sure there will be a number that is less than the square root that also divides n (the result of dividing the original number by the first will give you a number that is lower than the square root) so you only need to go up until the square root. While calculating the square root can be tedious (there's a library function, but let's go on), it is only done once, so it is a good point to use it. Also, you can initialy try dividing the number by two, and then skip all the even numbers, by adding 2 to j, instead of incrementing.
#include <math.h>
/* ... */
int prime(unsigned n) {
/* check for special cases */
if (n >= 1 && n <= 3) return TRUE; /* all these numbers are prime */
if (n % 2 == 0) return FALSE; /* all these numbers are not */
/* calculate (only once) the rounded down integer square root */
int j, square_root = isqrt(n); /* see below */
for (j = 3; j <= square_root; j += 2) { /* go two by two */
if (n % j == 0)
return FALSE;
}
/* if we reach here, all tests failed, so the number must be prime */
return TRUE;
}
While there's a sqrt() function in <math.h>, I recommend you to write an integer version of the square root routine (you can devise it easily) so you don't need to calculate it in full precision (just to integer precision).
/* the idea of this algorithm is that we have two numbers between 1 and n,
* the greater being the arithmetic mean between the previous two, while
* the lower is the result of dividing the original n by the arithmetic mean.
* it is sure than if we select the arithmetic mean, the number will be
* between the previous ones, and if I divide n by a number that is lower,
* the quotient will be higher than the original number. By the way, the
* arithmetic mean is always bigger than the square root, so the quotient
* will be smaller. At each step, both numbers are closer to each other, and
* so, the smaller is closer to the result of dividing n by itself (and this
* is the square root!)
*/
unsigned isqrt(unsigned n)
{
unsigned geom = 1, arith = n;
while (geom < arith) {
arith = (geom + arith) / 2;
geom = n / arith;
}
/* return the smaller of the two */
return arith;
}
so, your program would be:
#include <stdio.h>
#define FALSE (0)
#define TRUE (!FALSE)
unsigned isqrt(unsigned n)
{
unsigned geom = 1, arith = n;
while (geom < arith) {
arith = (geom + arith) / 2;
geom = n / arith;
}
return arith;
}
int prime(unsigned n) {
/* check for special cases */
if (n >= 1 && n <= 3) return TRUE;
if (n % 2 == 0) return FALSE;
/* calculate (only once) the rounded down integer square root */
int j, square_root = isqrt(n);
for (j = 3; j <= square_root; j += 2) {
if (n % j == 0) {
return FALSE;
}
}
return TRUE;
}
int main() {
unsigned n1, n2, i;
scanf("%u%u", &n1, &n2);
for (i = n1; i <= n2; i++) {
if (prime(i))
printf("%u\n", i);
}
return 0;
}
If you try your version against this one, with values like 2000000000 and 2000000100 you will see how this is saving a lot of calculations (indeed, for the cases below, the case of considering only the odd numbers when going throug the loop will take out of it half the numbers ---this is 1000000000 tests---, but the square root will reduce the number of tests to its square root ---only around 40000 tests--- for each number!!!).
$ primes
2000000000 2000000100
2000000011
2000000033
2000000063
2000000087
2000000089
2000000099
$ _
Your version takes (on my system) this execution time:
$ echo 2000000000 2000100000 | time primes0 >/dev/null
3.09user 0.00system 0:03.09elapsed 99%CPU (0avgtext+0avgdata 1468maxresident)k
0inputs+0outputs (0major+69minor)pagefaults 0swaps
$ _
while the version proposed takes:
$ echo 2000000000 2000100000 | time primes >/dev/null
0.78user 0.00system 0:00.78elapsed 99%CPU (0avgtext+0avgdata 1572maxresident)k
0inputs+0outputs (0major+72minor)pagefaults 0swaps
$ _

I want to optimize this program

I have recently started to learn c and as a programming exercise, I've written a program that computes and lists out prime numbers from 0 up to a maximum entered by the user. It's a rather short program so I'll post the source code here.
// playground.c
#include <stdio.h>
#include <stdbool.h>
#include <math.h>
int main ()
{
int max;
printf("Please enter the maximum number up to which you would like to see all primes listed: "
); scanf("%i", &max);
printf("All prime numbers in the range 0 to %i:\nPrime number: 2\n", max);
bool isComposite;
int primesSoFar[(max >> 1) + 1];
primesSoFar[0] = 2;
int nextIdx = 1;
for (int i = 2; i <= max; i++)
{
isComposite = false;
for (int k = 2; k <= (int)sqrt(i) + 1; k++)
{
if (k - 2 < nextIdx)
{
if (i % primesSoFar[k - 2] == 0)
{
isComposite = true;
k = primesSoFar[k - 2];
}
}else
{
if (i % k == 0) isComposite = true;
}
}
if (!isComposite)
{
printf("Prime number: %i\n", i);
primesSoFar[nextIdx] = i;
nextIdx++;
}
}
double primeRatio = (double)(nextIdx + 1) / (double)(max);
printf("The ratio of prime numbers to composites in range 0 to %d is %lf", max, primeRatio);
return 0;
}
I have become strangely fascinated with optimizing this program but I've hit a wall. The array primesSoFar is allocated based on a computed maximum size which ideally would be no larger than the number of prime numbers from 0 to max. Even if it were just slightly larger, that would be fine; as long as it's not smaller. Is there a way to compute the size the array needs to be that doesn't depend on first computing the primes up to max?
I've updated the code both applying suggested optimizations and adding internal documentation wherever it seemed helpful.
// can compute all the primes up to 0x3FE977 (4_188_535). Largest prime 4_188_533
#include <stdio.h>
#include <stdbool.h>
#include <math.h>
int main ()
{
int max;
printf("Please enter the maximum number up to which you would like to see all primes listed: "
); scanf("%i", &max);
// The algorithm proper doesn't print 2.
printf("All prime numbers in the range 0 to %i:\nPrime number: 2\n", max);
bool isComposite;
// primesSoFar is a memory hog. It'd be nice to reduce its size in proportion to max. The frequency
// of primes diminishes at higher numerical ranges. A formula for calculating the number of primes for
// a given numerical range would be nice. Sadly, it's not linear.
int PRIMES_MAX_SIZE = (max >> 1) + 1;
int primesSoFar[PRIMES_MAX_SIZE];
primesSoFar[0] = 2;
int nextIdx = 1;
int startConsecCount = 0;
for (int i = 2; i <= max; i++)
{
isComposite = false; // Assume the current number isn't composite.
for (int k = 2; k <= (int)sqrt(i) + 1; k++)
{
if (k - 2 < nextIdx) // Check it against all primes found so far.
{
if (i % primesSoFar[k - 2] == 0)
{
// If i is divisible by a previous prime number, break.
isComposite = true;
break;
}else
{
// Prepare to start counting consecutive integers at the largest prime + 1. if i
// isn't divisible by any of the primes found so far.
startConsecCount = primesSoFar[k - 2] + 1;
}
}else
{
if (startConsecCount != 0) // Begin counting consecutively at the largest prime + 1.
{
k = startConsecCount;
startConsecCount = 0;
}
if (i % k == 0)
{
// If i is divisible by some value of k, break.
isComposite = true;
break;
}
}
}
if (!isComposite)
{
printf("Prime number: %i\n", i);
if (nextIdx < PRIMES_MAX_SIZE)
{
// If the memory allocated for the array is sufficient to store an additional prime, do so.
primesSoFar[nextIdx] = i;
nextIdx++;
}
}
}
// I'm using this to get data with which I can find a way to compute a smaller size for primesSoFar.
double primeRatio = (double)(nextIdx + 1) / (double)(max);
printf("The ratio of prime numbers to composites in range 0 to %d is %lf\n", max, primeRatio);
return 0;
}
edit: primesSoFar should be half the size of the range 0 to max. No doubt that's caused some confusion.
I can give you two main ideas as I have worked on a project discussing this problem.
A prime number bigger than 3 is either 6k-1 or 6k+1, so for example 183 can't be prime because 183=6x30+3, so you don't even have to check it. (Be careful, this condition is necessary but not sufficient, 25 for exemple is 6x4+1 but is not prime)
A number is prime if it can't be divided by any prime number smaller or equal to its root, so it's preferable to take a benefit out of the smaller primes you already found.
Thus, you can start with a primesList containing 2 and 3, and iterate k to test all the 6k-1 and 6k+1 numbers (5, 7, 11, 13, 17, 19, 23, 25...) using the second rule I gave you, by using division on elements in the primesList which are smaller than or equal to the root of the number you are checking, if you found only one element dividing it, you just stop and pass to another element, 'cause this one is not prime, otherwise (if no one can divide it): update the primesList by adding this new prime number.
There is some debugging to be done first.
When I saw that the test was <= my brain said BUG as Arrays are subscripted from 0 .. max - 1.
for (int i = 2; i <= max; i++)
So I went to look at the array.
int primesSoFar[(max >> 1) + 1];
Oh he is adding one to the size so it should be ok.
Wait. Why is that shift in there? (max >> 1) is a divide by two.
I compiled the code and ran it, and MSVC reported a memory error.
I removed the shift, and the memory error report went away. The program worked as expected.
With that out of the way, PiNaKa30 and II Saggio Vecchino have very good advice. The choice of algorithm is going to effect the performance dramatically.
Mat gives very good advice. Read the Wikipedia entry. It is filled with wonderful information.
Picking the correct algorithm is key.
How you represent the data you are checking is a factor. int has a maximum value it can hold.
A performance profiler can tell you lots of useful information about where the Hot Spots are in your program.
Congratulations on your efforts in learning C. You picked a very good learning path.
The source code that follows is basically a rewrite. It's running now as I write this. I entered 0x7FFF_FFFF, the 32-bit signed integer positive maximum. In mere minutes on my Acer aspire laptop running on an AMD ryzen 3 with Linux Mint it's already in the hundreds of millions! The memory usage of the old version was half of max, rendering anything larger than 0x3EF977 impossible on my 4gb of RAM. Now it only uses 370728 bytes of memory for its array data when computing primes from 0 to 2_147_483_647.
/*
A super optimized prime number generator using my own implementation of the sieve of Eratosthenes.
*/
#include <stdio.h>
#include <stdbool.h>
#include <math.h>
int main ()
{
int max;
printf("Please enter the maximum to which you would like to see all primes listed: "
); scanf("%i", &max);
/*
Primes and their multiples will be stored until the next multiple of the prime is larger than max.
That prime and its corresponding multiple will then be replaced with a new prime and its corresponding
multiple.
*/
int PRIMES_MAX_SIZE = (int)sqrt(max) + 1;
int primes[PRIMES_MAX_SIZE];
int multiples[PRIMES_MAX_SIZE];
primes[0] = 2;
multiples[0] = 2;
int nextIdx = 1;
int const NO_DISPOSE_SENTINAL_VALUE = -1;
int nextDispose = NO_DISPOSE_SENTINAL_VALUE;
int startConsecCount = 0;
int updateFactor;
bool isComposite;
printf("All prime numbers in the range 0 to %i:\n\n", max);
// Iterate from i = 2 to i = max and test each i for primality.
for (int i = 2; i <= max; i++)
{
isComposite = false;
/*
Check whether the current i is prime by comparing it with the current multiples of
prime numbers, updating them when they are less than the current i and then proceeding
to check whether any consecutive integers up to sqrt(i) divide the current i evenly.
*/
for (int k = 2; k < (int)sqrt(i) + 1; k++)
{
if (k < nextIdx)
{
// Update the multiple of a prime if it's smaller than the current i.
if (multiples[k] < i)
{
updateFactor = (int)(i / primes[k]);
multiples[k] = updateFactor * primes[k] + primes[k];
// Mark the value for disposal if it's greater than sqrt(max).
if (multiples[k] > (int)sqrt(max)) nextDispose = k;
}
if (i == multiples[k])
{
isComposite = true;
break;
}else
{
startConsecCount = multiples[k] + 1;
}
} else
{
if (startConsecCount != 0)
{
k = startConsecCount;
startConsecCount = 0;
}
if (i % k == 0)
{
isComposite = true;
break;
}
}
}
/*
Print the prime numbers and either insert them at indices occupied by disposed primes or at
the next array index if available.
*/
if (!isComposite)
{
printf("Prime number: %i\n", i);
if (nextDispose != NO_DISPOSE_SENTINAL_VALUE)
{
primes[nextDispose] = i;
// This will trigger the update code before the comparison in the inner loop.
multiples[nextDispose] = 0;
nextDispose = NO_DISPOSE_SENTINAL_VALUE;
}else
{
if (nextIdx < PRIMES_MAX_SIZE)
{
primes[nextIdx] = i;
multiples[nextIdx] = 0;
}
}
}
}
return 0;
}
This thing will do the old 0 to 0x3EF977 in the blink of an eye. The old version couldn't do the 32-bit maximum on my system. It's on 201 million + already. I am super chuffed with the results. Thank you for your advice. I wouldn't have made it this far without help.

Creating a program in C to generate primes

I am trying to generate a list of all the prime numbers for the first 1000 numbers.
I am not sure where I am going wrong in my code. From what I can tell, my nested for loop is not reading dividing/reading the array correctly and then assigning that array the proper value. How can I fix it?
The program currently only generates all the odd numbers.
int main() {
int x = 1;
int arr[500];
int i, j, k;
int counter;
int primearray[500];
for (j = 0; j <= 500; j++) {
x += 2;
arr[j] = x;
for (k = 1; k <= 15; k++) {
counter = x % k;
if (counter == 0) {
primearray[j] = x;
} else {
break;
}
}
for (i = 0; i < 500; i++) {
printf("%d ", primearray[i]);
}
}
Please invest time in learning how to indent your code. Choose a style that suits you and use it consistently: this will make your programs easier to read, and in turn easier to understand.
As I'm writing this, your posted code doesn't even compile because a closing curly brace } is missing: such editing mistakes are made possible by misleading indentation. Also note that in a properly written C program you must remember to #include any standard headers that are used:
#include <stdio.h> // for `printf()`
Rather than try to fix your algorithm, which at first glance doesn't make any sense to me anyway, I will try to help you restructure your program.
Keep main() simple
Given that the goal of your program is checking which of the first 1000 natural numbers are prime, the main() function should do no more than loop through those numbers and print the ones which are prime, like this:
for (int n=0; n < 1000; ++n)
if (is_prime(n))
printf("%d\n", n);
Putting them in an array instead of printing is equally easy:
int prime_array[500]; // array of primes
int k=0; // current index in array of primes
for (int n=0; n < 1000; ++n)
if (is_prime(n))
prime_array[k++] = n;
Break up the program in several functions
In accordance to the previous idea, write short and simple functions that do one thing, and do it well. In your case, you should write the is_prime() function to determine if a number is prime or not. You can start from here:
///
/// #brief Checks if a number is prime.
/// #param [in] n Number to be checked
/// #returns Whether `n` is prime or not.
/// #retval 1 If `n` is prime.
/// #retval 0 If `n` is not prime.
///
int is_prime(int n)
{
// TODO: add code here
}
Decide how to check for primality
There is a Primality test article on Wikipedia that you should read.
First, you must correctly handle these special cases:
0 is not prime
1 is not prime
2 is prime
// TODO: also check 1 and 2 in a similar fashion
if (n == 0)
return 0;
After this is done, you can use a naive and inefficient algorithm that checks the other numbers:
// try divisors from 2 to n-1
for (int d=2; d < n; ++d)
if (n % d == 0) // if the division was even,
return 0; // the number is not prime
return 1; // if we get here, the number is prime
If you want to use a faster (but more complicated) algorithm for checking primes, look back at the Wikipedia article linked above. Notice how you'd only have to change the code inside is_prime() and the rest of the program would work the same, unchanged.
As I understood from your code, arr is an array of possible candidates and primearray is an array of approved ones. No every candidate will be approved one so you need different variables for indexing them.
The second issue is the algorithm for approving candidates. From this part of your code (I changed some indents)
for (k = 1; k <= 15; k++) {
counter = x%k;
if (counter == 0) {
primearray[j] = x;
} else {
break;
}
follows that you approve a candidate if it is divisible by all integers from 1 to 15 - I am sorry but prime numbers have not this property.
I think you could refer to this code which will generate all prime numbers up to the number that you specify. I think this will be more optimised.
void main()
{
int n, i, j, temp=0;
printf("Enter a number \n");
scanf("%d", &n);
printf(" Prime numbers -\n");
for(i=2; i<n+1; i++)
{
temp = 0;
for(j=2; j<i; j++)
{
if(i%j == 0)
{
temp = 1;
break;
}
}
if(temp == 0)
{
printf("%d \n", i);
}
}
getch();
}

SSE Programming in C — what to do for size not divisible by 4?

Here's the code which determines whether a matrix has orthogonal columns or not . The code is running properly when I have n which is divisible by 4, but when n is not divisible by 4, the program unexpectedly stops (mentioned position in the code)
for(i=0;i<n-1;i++)
{
for(j=i+1;j<n;j++)
{
sum = 0;
for(k=0;k<n-4;k+=4)
{
X=_mm_load_ps(&D[n*i+k]);
Y=_mm_load_ps(&D[n*j+k]);
printf("fff"); //not printing , program stops here
acc = _mm_add_ps(acc,_mm_mul_ps(X,Y));
}
_mm_store_ps(&temp[0],acc);
sum = temp[0]+temp[1]+temp[2]+temp[3];
for(;k<n;k++){
sum = sum + D[i*n+k]*D[j*n+k];
}
if(sum ==0)
return 1;
}
}
return 0;
}
What could be the possible reason for that? How do you handle arrays with size not divisible by 4?
Your code that determines when to stop early is not correct in the for(k...) loop. Try something like this:
int n4 = n - (n % 4); // smallest multiple of 4 <= n
for (k=0; k < n4, k+=4) //replaces: for(k=0;k<n-4;k+=4)
{ // loop body
}
for (k=n4; k < n; k++)
{ // non-vectorized loop for the last few columns
}

Resources