Description
This assignment is all about file systems. While completing this assignment,
1. You will gain additional practical experience writing code in the C programming language.
2. You will gain a better understanding of the implementation of at least one file system
(specifically, the MS-DOS file system that is used by most digital cameras and MP3 players).
The assignment is divided into three parts. The division is meant to provide guidelines
to help you allocate time when you are working on the assignment, and to help you make
steady progress on it instead of leaving it to the last minute. Although the assignment
is broken into three parts you are not required to handin the intermediate parts and will
handin only the completed assignment. For some people this is a challenging assignment,
so you should get started early. If you want you can work with one other person. Groups
of three or more people are not permitted.
The program you are to write, named fat12info is to be written from scratch using
C and it must compile and run on the undergraduate Linux machines provided by the
department. If you are you using your own machine make sure to test that it compiles and
runs as expected before handing it in. Keep in mind that the department machines are all
running a 64 bit version of Linux and as a result there could be differences in prototypes
provided by different include files and of course there are always the potential issues that
arise when porting from a machine with one native word size to another with a different
size. Also, you are not allowed to use C++ to complete this assignment. You are only
allowed to use functions from the standard C library. In particular, this means that you are
not allowed to use a library of functions that deal with FAT file systems from somewhere
on the Web.
When writing this program you are to use good software design practices for the naming
of functions, variables, and the organization of functions into separate files. In particular
this means that you are not to put all of your functions into a single file. Instead, put
related functions into the same file. For example, putting functions dealing with block I/O
together, and then putting directory functions together would make a lot of sense. To deal
with compiling these separate files into a single executable you will want to use a makefile.
The zip file associated with this assignment contains a suggested collection of files to place
your code, function prototypes, structure definitions etc., into along with a makefile. These
files are just suggestions and you are free to delete, rename, or add to the collection of files
that comprise this assignment.
1
Part 1
You should aim at completing this part of the assignment by Wednesday November 14th,
2012. The goal of this part is to help you become comfortable using random access techniques to access the data in a file and then interpreting some of the files contents. You
will find on the course web site a file fat volume.dat that contains the image of a small
MS-DOS file system. The task for this first part is to write code to output the following
information about this file system:
• its sector size.
• its cluster size in sectors.
• the number of entries in its root directory.
• the number of sectors per file allocation table.
• the number of reserved sectors on the disk.
• the number of hidden sectors on the disk.
• the sector number of the first copy of the file allocation table.
• the sector number of the first sector of the root directory.
• the sector number of the first sector of the first usable data cluster.
All of this information can be computed from the information stored in the first sector of
the file system. Full details about the MS-DOS file systems can be found here:
http://www.win.tue.nl/∼aeb/linux/fs/fat/fat-1.html
One of the sentences on this page is somewhat confusing. Under the heading “FAT12”, you
will find the statement:
Since 12 bits is not an integral number of bytes, we have to specify how these are
arranged. Two FAT12 entries are stored into three bytes; if these bytes are uv, wx, yz then
the entries are xuv and yzw.
Here is a simpler interpretation of the second sentence: Suppose you have the bytes b1,
b2 and b3 in that order. You can extract the two FAT12 entries by treating these three bytes
as a single little-endian 24 bit unsigned integer x. To get the FAT12 entry that occupies
the low-order 12 bits and x with 0x0FFF. To the get the FAT12 entry that occupies bits
12 to 23 bits shift the x value to the right by 12 so that bits 12 to 23 are now bits 0 to 11.
Unless you are 100% sure that the top order 8 bits were 0, and this value with 0x0FFF to
ensure that only the 12 bits of interest are captured.
When writing your code for part 1, be sure to make it general enough to use again for
parts 2 and 3. For instance it might be useful to define a struct like filesystem info that
contains most of the information about the file system that part 1 asks you to determine.
It is strongly recommend that you write functions to:
2
• Read a 512 byte sector into a buffer in memory since the approach of reading one byte
at a time from a file is extremely inefficient.
• Read a single cluster (In the DOS world the term cluster corresponds to the term block
that we have used in class) into a buffer in memory.
• Extract from a memory buffer an unsigned integer value from two (or three, or four)
specific bytes treated as a little-endian value.
You may find the manual pages for the C library functions open, lseek, and read
helpful. To access one of them, simply type
man 2 open
(for instance) in a terminal window1
.
Your test program must take a single argument, the name of the file containing the file
system it is supposed to interpret. That is, your main function should look at its parameters
argc and argv to determine which file should be used.
Part 2
You should aim at completing this part of the assignment by Thursday November 15, 2012.
The goal of this part is to be able to deal with directory entries, and extract information
about the files a directory contains. More specifically, you should add to your code from
part 1 so it can output the following information about each file in the root directory:
• Its name, including the extension if one is present.
• Whether or not this file is a directory.
• The number of the first cluster containing its data.
• Its size (note that directories have a size of 0).
This part of the assignment should be relatively easy once you have completed part 1.
Part 3
To complete this part, the program is to print out the information from part 2 about every
file in the file system, including files in subdirectories. The name printed for each file
is to include its full path name (for instance, ASS2/MCHEME.TXT). The program is also to
print out the list of clusters used for each file, in whatever format you prefer. This part of
the assignment is to be completed by adding to the solution from Part 2.
1
the 2 is necessary because there are several references to things called open, and you want the manual
page for the C library function, instead of the Tcl built-in command, the Perl pragma, or something else
3
This part of the assignment will likely be the most challenging because it deals directly
with the File Allocation Table. Note that because the file system is relatively small, it is
stored using the FAT12 format. It is recommended that you write functions that:
• determine if a given cluster number is the last cluster in its file.
• retrieve the next cluster number in a file, given the current cluster number.
Deliverables
You are to use the handin program to submit the assignment. The assignment name is a5,
and the files to submit for this part are:
1. All of your C source and include files. They must be commented appropriately, and
marks will be taken off it they do not contain enough comments to allow the TAs to figure
out what your code is doing.
2. A makefile such that when make is typed in the directory containing your solution the
program fat12info is built. This program takes a single argument, the name of the file
containing the sample file system the program is to decode. Your code must compile by
typing make and there must be no errors or warnings. You are not allowed to change the
gcc options in the makefile to alter how and when any warning messages are printed.
3. A file in text format that contains the following information:
• Your name and student number.
• The output from your program.
• How long it took you to complete this assignment.
• DO NOT handin a partner.txt file
If you are part of a group, only one of you are to submit the files as listed above. The
other member of the group must submit a single file called partner.txt that contains the
undergraduate login ID of your other partner (for instance, c3p0) and nothing else. This
will allow us to properly associate you with the work that was done with your partner.
Makefiles
In the batch of files for this assignment is included a Makefile to be used when building the
program. If you add any new files you will need to edit the Makefile so that the line that
starts “SRC =” contains the name of the new .c file. After modifying and saving the
makefile you will need to run makedepend to update the makefile with any .h dependencies.
Any time you add the include of a .h file to either a .c file or another .h file you need to
rerun makedepend. A more detailed description on how to use make and how makefiles
work can be found at:
4
http://www.ugrad.cs.ubc.ca/∼cs219/CourseNotes/Make/intro.html
5
