Pointers, Structs, and Memory Allocation

Lecture 11

• Pointers to functions implement polymorphism
• This changes the behavior, it can change at run-time which differs from static function calls.
• Without encapsulating the pointer name in parentheses for a function pointer, the compiler will mistake it.
• A struct is a user defined data type that groups different types of items into a single type
• The items inside a struct are known as members
• Values in structs are stored contiguously in memory
• Members of a structure can be directly accessed using the "." operator
• Memory can be allocated dynamically when the size is unknown at compile time
• This is useful for building data structures like linked lists, trees, and graphs
• When accessing members of a structure using pointers, the arrow operator (->) must be used
• It's more efficient to pass a pointer to a function versus the entire structure because only the memory address is passed
• Explicit memory allocation is faster than implicit memory allocation
• It uses less memory but throws more errors

Lecture 12

• Enums are a user defined data type that represent a group of global constants
• An example of an Enum is Days of the Week
• Using an Enum is better because it does not establish any type checking, and does not obey scope rules
• When using #define, related variables cannot be grouped
• The default value assigned to each member of the enum is 0
• Each value increases by 1
• For an enum Weekdays { MONDAY = 20, TUESDAY = 30, WEDNESDAY = 40, THURSDAY = 50, FRIDAY = 60 };
• TUESDAY = 30
• THURSDAY = 50
• A union is a user defined datatype consisting of a sequence of members whose storage overlaps
• Data types can be stored, but since it is all-shared memory space, only one member can hold a value at any given time.
• Only 1 member can hold a value at one time
• Size of a union is based on the size of its largest member
• Unions save space and memory, useful for plugins, slicing network packets

Lecture 13

• Members of a union can be accessed using the "." operator
• BASH, also known as the Bourne Again Shell, it's a command interpreter and a high-level programming language
• Scripting requires the shebang sequence of characters (#!) / (#!/bin/bash) on the first line
• This tells the operating system which shell should execute the file and follows the pathname
• When a script is created, its permissions must be updated so it can be read and executed
• Examples of permission combinations:
• 777 → Read, write, and execute for the owner, group, and others.
• 666 → Read and write for everyone
• 444 → Read-only for everyone
• BASH scripts can be expected to be analyzed
• Redirection encompasses the ways to alter where the standard input of a command comes from and where the standard output goes to
• Redirection uses the symbols < and >
• A file descriptor is a place where a program sends its output to and gets its input from
• There are 3 file descriptors: Standard input, Standard output, standard error.
• /etc/passwd is important because you can read and identify users in this file
• The actual passwords are NOT stored in /etc/passwd
• The first four fields of a /etc/passwd entry is for represtation

Lecture 14

• The steps needed to add a node to the front/tail of a list in a doubly linked list
• Doubly linked list using a struct

struct Node{
   int data;
   struct Node* next;
   struct Node* prev;
};
struct Node* createNode (int data) {
   struct Node* newNode = (struct Node*)malloc(sizeof(struct Node));
   newNode->data = data;
   newNode->next = NULL;
   newNode->prev = NULL;
   return newNode;
}

Lecture 14_3

• Header files include the #include <header.h> directive
• The C preprocessing directive #include is a request to the compiler
• Its purpose is to load the contents of a specific header file so that they can be used in the program
• Double inclusion is including the header file in multiple files
• The solution for double inclusion is to include the same header file twice or a Header Guard
• Header guards work by checking if it's been defined before.
• If it's the first time it won't be defined so then it will be ignored
• Linking is the process of collecting and combining various pieces of code and data into a single file that can be copied into memory and executed
• It can be done at compile time, load time, and run time
• Linking multiple files together leads to better optimization

Lecture 15

• Symbol resolution associates each symbol reference with exactly one symbol definition
• A symbol table is an array of structs
• Each entry contains the name, size, and location of the symbol.
• Relocation
• Three different types of object files:
• Relocatable object (.o file)
• Executable object file (a.out file)
• Shared object file (.so file)
• Global, external, and local symbols
• When dealing with duplicate symbols:
• Multiple strong symbols are not allowed
• Each item can only be defined once
• Otherwise, a Linker Error is created
• Given a strong symbol and multiple weak symbols, choose the strong symbol.
• If there are multiple weak symbols, pick an arbitrary one.

Lecture 16_2

• Putting all standard C functions into a single relocatable object module is bad because it is a huge waste of space
• A change would require recompilation of the entire source code.
• Static libraries are stored on disk in Archive files.
• The disadvantages of static libraries:
• Duplication in stored exe
• Duplication in running exe
• Minor bug fixes of system libraries require each application to explicitly relink.
• Shared Libraries, are the modern solution
• Dynamic linking can be loaded at runtime or load time
• Loading can be performed at an arbitrary memory address and linked with a program in memory.
• Shared libraries are "shared" in two ways:
• One .so file
• A single copy of the text section of a shared library in memory

Lecture 17

• The simplest form of control flow is Startup to Shutdown
• A CPU reads and executes a sequence of instructions, one at a time (CPU Control Flow)
• Events that cause abrupt changes in control on the system may include:
• Instruction divides by zero
• Data from disk or network adapter
• Control
• Hardware timer goes off
• These abrupt changes are referred to as Exceptional Control Flow (ECF)
• The three levels of the computer system where exceptional control flow can occur:
• The Hardware
• OS
• Application level
• An exception is the transfer of control to the OS kernel in response to some event (like a change in the processor state)
• An "event" is a significant change that occurs
• Types of exceptions: Interrupts, traps, faults, aborts
• A process is an instance of a program in execution
• In a logical control flow, processes take turns using the processor
• Each process executes a portion of its flow and then is preempted while other processes take their turns
• Preempted means that it's temporarily suspended
• A logical flow whose execution overlaps in time with another flow is called a concurrent flow

Lecture 18

• Init is is the first process in the system
• The kernel assigns a unique ID number to every process
• This ID is called a PID (Process ID)
• Parent-child relationship is formed between processes
• States a process may be in:
• Ready/Runnable
• Running
• Zombie
• Stopped
• Terminated
• Sleeping
• The process table contains all the information of the processes

Lecture 19

• A race condition is when threads can access shared data and try to change it at the same time
• A mutex lock, also known as a mutual exclusion, is an operation used to prevent simultaneous possession of the shared resources.
• Critical section is a portion of the code only one thread should execute at a time
• Semaphores are used to coordinate access to resources
• Binary and Counting are the two types of semaphores
• The functions that lock/unlock a semaphore are Sem_post(), sem_wait()

Lecture 20

• A signal is a small message that notifies a process of an event
• A signal may be sent for sending or receiving purposes
• When receiving a process, the process can do one of three things
• Ignore
• Catch
• Terminate
• A pending signal has been sent but not yet received
• A process can selectively block the receipt of certain signals
• Since there can be at most one pending signal of a particular type, the rest of the signals that may be of the same type, gets discarded
• Each signal type has a predefined default action:
• The Process terminates
• The Process terminates and dumps core.
• The process stops (suspends) until restarted by a SIGCONT signal.
• The process ignores the signal
• Steps needed for a transaction between a client and a server:
• The client initiates a request to the server
• The server is like OK what is this what are you looking for then stamps it for approval
• The server sends a response to the client and waits
• The client receives the response, then processes it
• A network is a hierarchical system organized by geographical proximity
• SAN, LAN, WAN, are three proximities
• IP addresses are stored in memory for a network
• Referred to as Dotted Decimal notation
• A socket is an endpoint of a connection
• A socket pair is indicated as (IP_adress:port)
• An ephemeral port assigned automatically by client kernel when a client makes a connection request
• A well-known port is associated with some service provided by a server
• Converting byte order between the network and the client is necessary
• Be able to fill in aspects of the socket address or the socket pair:
• AF_INET is needed
• SOCK_STREAM
• Client, server, and socket file descriptors
• Functions implement so the server can communicate with a client

Lecture 21

• The main objective of multiprogramming is to to have some process always running, maximizing CPU utilization
• CPU sitting idle is a waste, no work is being accomplished
• When one process waits, the OS gives the CPU to another process.
• The CPU scheduler selects a process from the processes in memory that are ready to execute and allocates the CPU to that process.
• The process control block shows all the information stored
• A preemptive scheme is when the OS can interrupt a running process and switch to another process before the current one finishes
• A nonpreemptive scheme is when scheduling takes place until the process is finished.
• Able to create the gantt chart and calculate the average wait time using each process
• Use different approaches like First come first serve, Shortest job first (preemptive and nonpreemptive), Round robin
• Calculate exponential averaging