I was trying to solve this problem - link. Segment Tree with Lazy Propagation. But I dont know where I made mistake. Please help me find the bug.
I am new to Segment Tree with Lazy Propagation. But my code seems ok.
#include <bits/stdc++.h>
using namespace std;
const int MAXN = 100000;
struct info{ long long sum, prop; }; info tree[300010];
void update(int node, int l, int r, int i, int j, int val) {
if( i > r || j < l) return;
if(i <= l && j >= r) {
tree[node].sum += (r-l+1)*val;
tree[node].prop += val;
return;
} int left = node*2, right = left|1, mid = (l+r)/2;
update(left, l, mid, i,j,val);
update(right, mid+1, r, i,j,val);
tree[node].sum = tree[left].sum + tree[right].sum + (r-l+1)*tree[node].prop;
}
long long query(int node, int l, int r, int i, int j, long long carry = 0) {
if(i > r || j < l ) return 0;
if(i <= l && j >= r) return tree[node].sum + (r-l+1)*carry;
int mid = (l+r)/2, left = node*2, right = left|1;
long long ret = query(left, l, mid, i,j, carry + tree[node].prop);
ret += query(right,mid+1,r,i,j, carry + tree[node].prop);
return ret;
}
int main(int argc, char const *argv[]) {
#ifndef ONLINE_JUDGE
freopen("in", "r", stdin);
#endif
int t,co=0; scanf("%d", &t); while(t--) {
int n, q; scanf("%d %d", &n, &q);
for(int i=0; i<=3*n; i++) tree[i].sum = tree[i].prop = 0;
printf("Case %d:\n", ++co);
while(q--) {
int type,a,b,c; scanf("%d", &type); if(!type) {
scanf("%d %d %d", &a, &b, &c);
update(1, 0, n-1, a,b,c);
} else {
scanf("%d %d", &a, &b);
printf("%d\n", query(1, 0, n-1, a,b));
}
}
}
}
First of all this site is not for finding bug in code. Try https://codereview.stackexchange.com/ for this purpose.
Anyway In your code return type of function query() is long long but in main you are printing integer type. change printf("%d\n", query(1, 0, n-1, a,b)); to printf("%lld\n", query(1, 0, n-1, a,b)); and I hope you will get AC.
Related
I use pthreads on ubuntu to implement multithreaded matrix-vector multiplication, but the runtime reports an error Segmentation fault
#pragma comment(lib, "pthreadVC2.lib")
#define _CRT_SECURE_NO_WARNINGS 1
#include <stdio.h>
#include <stdlib.h>
#include <pthread.h>
/* Global variables */
int thread_count = 8;
int m, n;
double* A = NULL;
double* x = NULL;
double* y = NULL;
/* Serial functions */
void Usage(char* prog_name);
void Read_matrix(char* prompt, double A[], int m, int n);
void Read_vector(char* prompt, double x[], int n);
void Print_matrix(char* title, double A[], int m, int n);
void Print_vector(char* title, double y[], double m);
/* Parallel function */
void* Pth_mat_vect(void* rank);
/*------------------------------------------------------------------*/
int main(int argc, char* argv[]) {
long thread;
pthread_t* thread_handles;
thread_count = atoi(argv[1]);
thread_handles = malloc(thread_count * sizeof(pthread_t));
printf("Enter m and n\n");
scanf("%d%d", &m, &n);
A = malloc(m * n * sizeof(double));
x = malloc(n * sizeof(double));
y = malloc(m * sizeof(double));
Read_matrix("Enter the matrix", A, m, n);
Print_matrix("We read", A, m, n);
Read_vector("Enter the vector", x, n);
Print_vector("We read", x, n);
for (thread = 0; thread < thread_count; thread++)
pthread_create(&thread_handles[thread], NULL,
Pth_mat_vect, (void*)thread);
for (thread = 0; thread < thread_count; thread++)
pthread_join(thread_handles[thread], NULL);
Print_vector("The product is", y, m);
free(A);
free(x);
free(y);
return 0;
} /* main */
/*------------------------------------------------------------------
* Function: Read_matrix
* Purpose: Read in the matrix
* In args: prompt, m, n
* Out arg: A
*/
void Read_matrix(char* prompt, double A[], int m, int n) {
int i, j;
printf("%s\n", prompt);
for (i = 0; i < m; i++)
for (j = 0; j < n; j++)
scanf("%lf", &A[i * n + j]);
} /* Read_matrix */
/*------------------------------------------------------------------
* Function: Read_vector
* Purpose: Read in the vector x
* In arg: prompt, n
* Out arg: x
*/
void Read_vector(char* prompt, double x[], int n) {
int i;
printf("%s\n", prompt);
for (i = 0; i < n; i++)
scanf("%lf", &x[i]);
} /* Read_vector */
/*------------------------------------------------------------------
* Function: Pth_mat_vect
* Purpose: Multiply an mxn matrix by an nx1 column vector
* In arg: rank
* Global in vars: A, x, m, n, thread_count
* Global out var: y
*/
void* Pth_mat_vect(void* rank) {
long my_rank = (long)rank;
int i, j;
int local_m = m / thread_count;
int my_first_row = my_rank * local_m;
int my_last_row = (my_rank + 1) * local_m - 1;
for (i = my_first_row; i <= my_last_row; i++) {
y[i] = 0.0;
for (j = 0; j < n; j++)
y[i] += A[i * n + j] * x[j];
}
return NULL;
} /* Pth_mat_vect */
/*------------------------------------------------------------------
* Function: Print_matrix
* Purpose: Print the matrix
* In args: title, A, m, n
*/
void Print_matrix(char* title, double A[], int m, int n) {
int i, j;
printf("%s\n", title);
for (i = 0; i < m; i++) {
for (j = 0; j < n; j++)
printf("%4.1f ", A[i * n + j]);
printf("\n");
}
} /* Print_matrix */
/*------------------------------------------------------------------
* Function: Print_vector
* Purpose: Print a vector
* In args: title, y, m
*/
void Print_vector(char* title, double y[], double m) {
int i;
printf("%s\n", title);
for (i = 0; i < m; i++)
printf("%4.1f ", y[i]);
printf("\n");
} /* Print_vector */
This code is from An Introduction to Parallel Programming
I know this error seems to be related to memory, in fact the code runs without entering main().I tried some other people's methods, but none of them worked.
I need to make a c program that will make a histogram of all the letters present in a phrase the user gives. When I run it, I does it but gives a "* stack smashing detected *: terminated". Where would this error be coming from? (for ease right now I set max to 3). In the future i'll have it find the max
Thank you
Andrew
#include <stdio.h>
#include <ctype.h>
#include <string.h>
static void ReadText(int histo[26],int max) {
char phrase[100];
int i;
char Letter;
char toArray;
// read in phrase
printf("Enter Phrase: "); // reads in phrase with spaces between words
scanf("%[^\n]",phrase);
// count the number of certain letters that occur
for(i = 0; i <= strlen(phrase);++i) {
Letter = phrase[i];
if(isalpha(Letter) != 0){
Letter = tolower(Letter);
toArray = Letter - 97;
histo[(int)toArray] = histo[(int)toArray] + 1;
}
}
}
static void DrawHist(int histo[26], int max){
int i;
int j;
int histo2[50];
for(i = 0; i <= 26; i++) {
histo2[i+i] = histo[i];
if(i < 25) {
histo2[i+i+1] = 0;
}
}
// (i = 1; i <= 50; i++) {
// printf("%d",histo2[i]);
//}
//printf("\n");
for(i=max;i>0;--i) {
for(j=0;j<=51;++j) {
if((j < 51) && (histo2[j] >= i)) {
printf("|");
}
else if((j < 51) && (histo2[j] < i)){
printf(" ");
}
else if(j == 51){
printf("\n");
}
}
}
printf("+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-\n");
printf("A B C D E F G H I J K L M N O P Q R S T U V W X Y Z\n");
}
int main() {
int histo[26] = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0};
int max = 3;
//int i;
ReadText(histo,max);
//for(i = 0; i<26;++i) {
// printf("%d",histo[i]);
//}
DrawHist(histo,max);
return 0;
}
Could anyone please help me understand why Clang reallocates the same memory address for different variables while their lifetimes intersect?
I am using a sample program (below) to show the problem.
When I compile the program with clang -O0, variable j in function ok has the same memory address as variable solutions in function nqueens.
Function ok is called inside function nqueens, which means that the lifetime of the variables intersect; the same stack space cannot be used/reused for both functions.
Compiling the program with gcc or clang at -O1, however, they are assigned different memory addresses.
Any help is appreciated!
#include <stdlib.h>
#include <stdio.h>
#include <memory.h>
#include <alloca.h>
/* Checking information */
static int solutions[] = {
1,
0,
0,
2,
10, /* 5 */
4,
40,
92,
352,
724, /* 10 */
2680,
14200,
73712,
365596,
};
#define MAX_SOLUTIONS sizeof(solutions)/sizeof(int)
int total_count;
int sharedVar = 0;
int ok(int n, char *a)
{
int i, j;
char p, q;
printf("jjjjjjjjj: %d, %p\n", n,&j);
for (i = 0; i < n; i++) {
p = a[i];
for (j = i + 1; j < n; j++) {
q = a[j];
if (q == p || q == p - (j - i) || q == p + (j - i))
return 0;
}
}
return 1;
}
void nqueens (int n, int j, char *a, int *solutions)
{
int i,res;
sharedVar = sharedVar * j - n;
if (n == j) {
/* good solution, count it */
*solutions = 1;
return;
}
printf("solutions: %d, %p\n", j, &solutions);
*solutions = 0;
/* try each possible position for queen <j> */
for (i = 0; i < n; i++) {
a[j] = (char) i;
if (ok(j + 1, a)) {
nqueens(n, j + 1, a,&res);
*solutions += res;
}
}
}
int main()
{
int size = 3;
char *a;
// printf("total_count: %p\n", &total_count);
total_count=0;
a = (char *)alloca(size * sizeof(char));
printf("Computing N-Queens algorithm (n=%d) ", size);
sharedVar = -5;
nqueens(size, 0, a, &total_count);
printf("completed!\n");
printf("sharedVar: %d\n", sharedVar);
}
I get _CrtIsValidHeapPointer(pUserData) error when running the code above.
Sometimes the code works perfectly, and sometimes this message appears. So I guess the problem is related to the memory allocation. But I've gone through the code many times and the numbers make sence to me (and also when debugging).
I noticed it happens in line "free(str_temp)" at the debugging.
The relevant code is here:
int main(){
int n;
int len;
char *str;
char command[3];
printf("Enter your string:\n");
scanf("%d", &n);
str = malloc(n+1);
scanf("%s", str);
while (1){
printf(">");
scanf("%s", command);
if (compare(command, "ml")) {
int k;
scanf("%d", &k);
multiply(str, n, k);
printf("Current string is %s\n", str);
n = ln(str);
continue;
}
free(str);
return 0;
}
void multiply(char *str, int n, int k) {
char *str_temp = malloc(n+1);
int i;
int j;
int q;
for (i = 0; i < n; i++){
str_temp[i] = str[i];
}
str_temp[n] = '\0';
free(str);
*str = malloc(n*k+1);
for (i = 0; i < k; i++){
for (j = 0; j < n; j++){
str[i*n + j] = str_temp[j];
}
}
str[n*k] = '\0';
free(str_temp);
}
Try to use message defination
void multiply(char **str, int n, int k)//Use **str(double pointer) instead of *str.
And call it like
multiply(&str, n, k);
I'm writing a code for the image denoising and came across a strange problem with stripes in the processed images. Basically when I'm calculating X-gradient of image the horizontal stripes appear (or vertical for Y direction) Lena X gradient.
The whole algorithm works OK and it looks like I'm getting the correct answer (I'm comparing with program in C) except those annoying stripes Lena result.
The distance between stripes is changing with different block sizes. I'm also having different stripes positions each time I run the program! Here is the part of the program related to the gradient calculation. I have a feeling that I'm doing something very stupid :) Thank you!
#define BLKXSIZE 16
#define BLKYSIZE 16
#define idivup(a, b) ( ((a)%(b) != 0) ? (a)/(b)+1 : (a)/(b) )
void Diff4th_GPU(float* A, float* B, int N, int M, int Z, float sigma, int iter, float tau, int type)
{
float *Ad;
dim3 dimBlock(BLKXSIZE,BLKYSIZE);
dim3 dimGrid(idivup(N,BLKXSIZE), idivup(M,BLKYSIZE));
cudaMalloc((void**)&Ad,N*M*sizeof(float));
cudaMemcpy(Ad,A,N*M*sizeof(float),cudaMemcpyHostToDevice);
cudaCheckErrors("cc1");
int n = 1;
while (n <= iter) {
Diff4th2D<<<dimGrid,dimBlock>>>(Ad, N, M, sigma, iter, tau, type);
n++;
cudaDeviceSynchronize();
cudaCheckErrors("kernel");}
cudaMemcpy(B,Ad,N*M*sizeof(float),cudaMemcpyDeviceToHost);
cudaCheckErrors("cc2");
cudaFree(Ad);
}
__global__ void Diff4th2D(float* A, int N, int M, float sigma, int iter, float tau, int type)
{
float gradX, gradX_sq, gradY, gradY_sq, gradXX, gradYY, gradXY, sq_sum, xy_2, Lam, V_norm, V_orth, c, c_sq, lam_t;
int i = blockIdx.x*blockDim.x + threadIdx.x;
int j = blockIdx.y*blockDim.y + threadIdx.y;
int index = j + i*N;
if ((i < N) && (j < M))
{
float gradX = 0, gradY = 0, gradXX = 0, gradYY = 0, gradXY = 0;
if ((i>1) && (i<N)) {
if ((j>1) && (j<M)){
int indexN = (j)+(i-1)*(N);
if (indexN > ((N*M)-1)) indexN = (N*M)-1;
if (indexN < 0) indexN = 0;
int indexS = (j)+(i+1)*(N);
if (indexS > ((N*M)-1)) indexS = (N*M)-1;
if (indexS < 0) indexS = 0;
int indexW = (j-1)+(i)*(N);
if (indexW > ((N*M)-1)) indexW = (N*M)-1;
if (indexW < 0) indexW = 0;
int indexE = (j+1)+(i)*(N);
if (indexE > ((N*M)-1)) indexE = (N*M)-1;
if (indexE < 0) indexE = 0;
gradX = 0.5*(A[indexN]-A[indexS]);
A[index] = gradX;
}
}
}
}
You have a race condition inside your kernel, as elements of A may or may not be overwritten before they are used.
Use different arrays for input and output.