From: VGoncalo <vitor.goncalo.costa@gmail.com>
Date: Sat, 14 Mar 2026 00:51:24 +0000 (+0000)
Subject: blocks strategy
X-Git-Url: https://vgcfreebox.myrthtech.pt/gitweb/ue-pp-squarematrixparallelisation.git/commitdiff_plain/45a840c39a529958bf07358750f988c758aad721?ds=inline;hp=3d95624a0cd7feb3992ef3ac38fbcfba8b75ae30

blocks strategy
---

diff --git a/BlocksParallelMatrixSearch.c b/BlocksParallelMatrixSearch.c
new file mode 100644
index 0000000..d2f91b2
--- /dev/null
+++ b/BlocksParallelMatrixSearch.c
@@ -0,0 +1,116 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <pthread.h>
+#include <time.h>
+#include <stdatomic.h>
+#include <math.h>
+
+const int m_rows = 40000;
+const int m_cols = m_rows;
+
+_Atomic int hv = 0;
+int **matrix;
+
+typedef struct {
+    int start_r;
+    int end_r;
+    int start_c;
+    int end_c;
+} thread_args;
+
+
+void* parallel_search_blocks(void *arg) {
+    thread_args *args = (thread_args *)arg;
+    for (int r = args->start_r; r < args->end_r; r++) {
+        for (int c = args->start_c; c < args->end_c; c++) {
+            if (matrix[r][c] > hv){
+                hv = matrix[r][c];
+            }
+        }
+    }
+    return NULL;
+}
+
+void run_parallel_search(int num_threads){
+    pthread_t *threads = malloc(num_threads * sizeof(pthread_t));
+    thread_args *args = malloc(num_threads * sizeof(thread_args));
+
+    if (threads == NULL || args == NULL) {
+        perror("Failed to allocate memory for threads");
+        exit(1);
+    }
+
+    int blocks_per_side = (int)sqrt(num_threads);
+    int row_chunk = m_rows / blocks_per_side;
+    int col_chunk = m_cols / blocks_per_side;
+
+    int t_count = 0;
+    for (int i = 0; i < blocks_per_side; i++) {
+        for (int j = 0; j < blocks_per_side; j++) {
+            args[t_count].start_r = i * row_chunk;
+            args[t_count].end_r = (i == blocks_per_side - 1) ? m_rows : (i + 1) * row_chunk;
+            
+            args[t_count].start_c = j * col_chunk;
+            args[t_count].end_c = (j == blocks_per_side - 1) ? m_cols : (j + 1) * col_chunk;
+
+            if (pthread_create(&threads[t_count], NULL, &parallel_search_blocks, &args[t_count]) != 0) {
+                perror("Failed to create thread");
+                exit(1);
+            }
+            t_count++;
+        }
+    }
+
+    for (int i = 0; i < t_count; i++) {
+        pthread_join(threads[i], NULL);
+    }
+    
+    free(threads);
+    free(args);
+}
+
+int main(void) {
+    printf("1st --> Allocate memory\n");
+    clock_t t_t1 = clock();
+    
+    matrix = malloc(m_rows * sizeof(int*));
+    if (matrix == NULL) { perror("malloc failed"); return 1; }
+    
+    int *data;
+    size_t data_size = (size_t)m_rows * m_cols * sizeof(int);
+    if (posix_memalign((void**)&data, 64, data_size) != 0) {
+        perror("alignment failed");
+        free(matrix);
+        return 1;
+    }
+    
+    t_t1 = clock() - t_t1;
+    printf(" %f sec\n", ((double)t_t1) / CLOCKS_PER_SEC);
+
+    printf("2nd --> Populate the matrix\n");
+    clock_t t_t2 = clock();
+    for (int r = 0; r < m_rows; r++) {
+        matrix[r] = data + r * m_cols;
+    }
+    for (int r = 0; r < m_rows; r++) {  
+        for (int c = 0; c < m_cols; c++) {
+            matrix[r][c] = r * m_cols + c;
+        }
+    }
+    t_t2 = clock() - t_t2;
+    printf(" %f sec\n", ((double)t_t2) / CLOCKS_PER_SEC);  
+  
+    printf("3rd --> Search matrix (Block Split)\n");
+    clock_t t_t3 = clock();
+    
+    int threads_to_use = 8000; 
+    run_parallel_search(threads_to_use);
+    
+    t_t3 = clock() - t_t3;
+    printf(" %f sec\n", ((double)t_t3) / CLOCKS_PER_SEC);
+ 
+    printf("Done\n The biggest value found is %d\n", hv); 
+    free(data);
+    free(matrix);
+    return 0;
+}
\ No newline at end of file
diff --git a/MatrixParallelPrograming.pdf b/MatrixParallelPrograming.pdf
index f2d04ae..76e014b 100644
Binary files a/MatrixParallelPrograming.pdf and b/MatrixParallelPrograming.pdf differ