From f0e969bd2e954e31d6750cefb8691659248f7746 Mon Sep 17 00:00:00 2001
From: Santo Cariotti <santo@dcariotti.me>
Date: Tue, 6 Jul 2021 16:25:30 +0200
Subject: os: add homework 3

---
 Year_2/OS/hw/hw3.c | 100 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 100 insertions(+)
 create mode 100644 Year_2/OS/hw/hw3.c

(limited to 'Year_2/OS/hw')

diff --git a/Year_2/OS/hw/hw3.c b/Year_2/OS/hw/hw3.c
new file mode 100644
index 0000000..b97543a
--- /dev/null
+++ b/Year_2/OS/hw/hw3.c
@@ -0,0 +1,100 @@
+/*
+    Homework n.3
+
+    Usando la possibilita' di mappare file in memoria, creare un programma che
+    possa manipolare un file arbitrariamente grande costituito da una sequenza
+    di record lunghi N byte.
+    La manipolazione consiste nel riordinare, tramite un algoritmo di ordinamento
+    a scelta, i record considerando il contenuto dello stesso come chiave:
+    ovvero, supponendo N=5, il record [4a a4 91 f0 01] precede [4a ff 10 01 a3].
+    La sintassi da supportare e' la seguente:
+     $ homework-3 <N> <pathname del file>
+
+    E' possibile testare il programma sul file 'esempio.txt' prodotto dal seguente
+    comando, utilizzando il parametro N=33:
+     $ ( for I in `seq 1000`; do echo $I | md5sum | cut -d' ' -f1 ; done ) > esempio.txt
+
+    Su tale file, l'output atteso e' il seguente:
+     $ homework-3 33 esempio.txt
+     $ head -n5 esempio.txt
+        000b64c5d808b7ae98718d6a191325b7
+        0116a06b764c420b8464f2068f2441c8
+        015b269d0f41db606bd2c724fb66545a
+        01b2f7c1a89cfe5fe8c89fa0771f0fde
+        01cdb6561bfb2fa34e4f870c90589125
+     $ tail -n5 esempio.txt
+        ff7345a22bc3605271ba122677d31cae
+        ff7f2c85af133d62c53b36a83edf0fd5
+        ffbee273c7bb76bb2d279aa9f36a43c5
+        ffbfc1313c9c855a32f98d7c4374aabd
+        ffd7e3b3836978b43da5378055843c67
+*/
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+
+/* this is an easy easy bubble sort algorithm */
+void sort(char* map, int n, int size)
+{
+    int lines = size / n;
+    int i, j;
+
+    char* tmp = malloc(n);
+    char *pi, *pj;
+
+    for (i = 0; i < lines; ++i) {
+        for (j = 0; j < lines - 1; ++j) {
+            pi = map + n * i;
+            pj = map + n * j;
+
+            if (memcmp(pi, pj, n) < 0) {
+                memcpy(tmp, pi, n);
+                memcpy(pi, pj, n);
+                memcpy(pj, tmp, n);
+            }
+        }
+    }
+
+    free(tmp);
+}
+
+int main(int argc, char** argv)
+{
+    int fd;
+    char* addr;
+    struct stat sb;
+    int w; // length of the line, w = width(?)
+
+    if (argc < 3) {
+        fprintf(stderr, "Use: %s <N> <file>", argv[0]);
+        exit(1);
+    }
+
+    if ((fd = open(argv[2], O_RDWR)) == -1) {
+        perror(argv[2]);
+        exit(1);
+    }
+
+    w = atoi(argv[1]);
+
+    if (fstat(fd, &sb) == -1) {
+        perror("fstat");
+        exit(1);
+    }
+
+    if ((addr = mmap(NULL, sb.st_size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0)) == MAP_FAILED) {
+        perror("mmap");
+        exit(1);
+    }
+    close(fd);
+
+    sort(addr, w, sb.st_size);
+
+    munmap(addr, sb.st_size);
+    return 0;
+}
-- 
cgit v1.2.3-18-g5258