csci5103_project3

File Systems

• Bryan Baker - [email protected]
• Alice Anderegg - [email protected]
• Hailin Archer - [email protected]

Assumptions

• The write function needs to work only as intended by the shell program.
• The file will be written and not updated.
• The disk map used to track the locations of all free data blocks can be any type of data structure. A byte array has been chosen.

How To Run

To build the files use make To create a disk image run the command: ./simplefs image.xxx xxx where xxx is the number of blocks you would like to create in the disk image. Execute commands to the shell program to interact with the file system. Use help to see a list of possibilities.

Function Definitions

Main Functions (Required)

• fs_format:

  • Purpose: Format the created disk image to use as a file system.
  • Input: None.
  • Output: a formatted disk image.
  • Return Value: 1 if successful, 0 otherwise.
  • Pseudo Code:
    • Check if mounted
    • Check if already formatted
    • Set and write superblock
    • Zero out the rest of the blocks of the file system.

• fs_debug:

  • Purpose: Print out the information of the file system. The amount of inodes possible, the number of blocks, and the data for each valid inode in the file system.
  • Input: None.
  • Output: A screen printout of file system information.
  • Return Value: void.
  • Pseudo Code:
    • Check if formatted
    • Print the number of blocks, inode blocks, and inodes
    • Print information for each valid inode: the size, the direct data blocks used, the indirect data block, and the indirect data blocks used.

• fs_mount:

  • Purpose: Mount the disk image to use as a file system
  • Input: None.
  • Output: A mounted disk image for a file system.
  • Return Value: 1 if successful, 0 otherwise.
  • Pseudo Code:
    • Check if mounted
    • Check if formatted
    • Read the data on the disk image and create a disk map

• fs_create:

  • Purpose: Create an inode.
  • Input: None.
  • Output: A valid inode will be created on the file system to prepare to write.
  • Return Value: The created inode number, -1 otherwise.
  • Pseudo Code:
    • Check if mounted
    • Check if inode table is full
    • Find a free inode slot
    • Write meta data to inode
    • Return the created inode number

• fs_delete:

  • Purpose: Remove an inode from the file system.
  • Input: The inode number of delete.
  • Output: The inode will be removed from the file system.
  • Return Value: 1 if successful, 0 otherwise.
  • Pseudo Code:
    • Check if mounted
    • Check if inode number is less than the valid range
    • Check if the inode number is valid
    • Write empty blocks to the direct nodes and update the disk map.
    • Check for indirect data and zero out if it exists. Update the disk map.
    • Remove the inode number and save the information.

• fs_getsize:

  • Purpose: Get the amount of data associated with an inode.
  • Input: An inode number.
  • Output: The inode size.
  • Return Value: An integer for the inode size.
  • Pseudo Code:
    • Check if mounted
    • Check is inode number is valid
    • Read and return the size of the inode.

• fs_read:

  • Purpose: Read data from the file system.
  • Input: The inode number, the data buffer, the length of data to read, and the offset.
  • Output: The data read from the file system.
  • Return Value: The number of bytes read, 0 otherwise.
  • Pseudo Code:
    • Check if mounted
    • Check if inode number is valid
    • Determine which blocks and bytes to start reading from
    • Read through data starting with the direct nodes and then on to the indirect nodes if required
    • Copy to output buffer
    • Return the number of bytes read

• fs_write:

  • Purpose: Write data to an inode in the file system.
  • Input: The inode number, the data buffer, the length of the data to write, and the offset.
  • Output: The data will be written in to the file system.
  • Return Value: The number of bytes written, 0 otherwise.
  • Pseudo Code:
    • Check if mounted
    • Check if inode number is valid
    • Determine which blocks to start writing to.
    • Find a free block to write to - return zero if disk is full
    • Find the number of bytes to write and read them from the buffer to a data block.
    • Write the data block to the disk and track using direct or indirect nodes.
    • Update the disk map
    • Update the inode size
    • Save the inode meta data
    • Return the number of bytes written

Helper Functions (created by the team)

• inode_load:

  • Purpose: Find an inode block using an inode number.

• inode_save:

  • Purpose: Save an inode block using an inode number.

• get_inode_cnt:

  • Purpose: Get a count of all of the valid inodes on the file system.

• find_free_block:

  • Purpose: Returns the value of a free block which can be used to write data.