1 @node Project 4--File Systems, References, Project 3--Virtual Memory, Top
2 @chapter Project 4: File Systems
4 In the previous two assignments, you made extensive use of a
5 filesystem without actually worrying about how it was implemented
6 underneath. For this last assignment, you will fill in the
7 implementation of the filesystem. You will be working primarily in
8 the @file{filesys} directory.
10 You should build on the code you wrote for the previous assignments.
11 However, if you wish, you may turn off your VM features, as they are
12 not vital to making the filesystem work. (You will need to edit
13 @file{filesys/Makefile.vars} to fully disable VM.) All of the
14 functionality needed for project 2 (argument passing, syscalls and
15 multiprogramming) must work in your filesys submission.
17 On the other hand, one of the particular charms of working on
18 operating systems is being able to use what you build, and building
19 full-featured systems. Therefore, you should strive to make all the
20 parts work together so that you can run VM and your filesystem at the
21 same time. Plus, keeping VM is a great way to stress-test your
22 filesystem implementation.
24 Your submission should define @code{THREAD_JOIN_IMPLEMENTED} in
25 @file{constants.h} (@pxref{Conditional Compilation}).
28 * File System New Code::
29 * Problem 4-1 Large Files::
30 * Problem 4-2 File Growth::
31 * Problem 4-3 Subdirectories::
32 * Problem 4-4 Buffer Cache::
33 * File System Design Document Requirements::
37 @node File System New Code
40 Here are some files that are probably new to you. These are in the
41 @file{filesys} directory except where indicated:
45 Simple utilities for the filesystem that are accessible from the
50 Top-level interface to the file system.
54 Translates file names to inodes. The directory data structure is
59 Manages the data structure representing the layout of a
64 Translates file reads and writes to disk sector reads
67 @item lib/kernel/bitmap.h
68 @itemx lib/kernel/bitmap.c
69 A bitmap data structure along with routines for reading and writing
70 the bitmap to disk files.
73 Our file system has a Unix-like interface, so you may also wish to
74 read the Unix man pages for @code{creat}, @code{open}, @code{close},
75 @code{read}, @code{write}, @code{lseek}, and @code{unlink}. Our file
76 system has calls that are similar, but not identical, to these. The
77 file system translates these calls into physical disk operations.
79 All the basic functionality is there in the code above, so that the
80 filesystem is usable right off the bat. In fact, you've been using it
81 in the previous two projects. However, it has severe limitations
82 which you will remove.
84 While most of your work will be in @file{filesys}, you should be
85 prepared for interactions with all previous parts (as usual).
87 @node Problem 4-1 Large Files
88 @section Problem 4-1: Large Files
90 Modify the file system to allow the maximum size of a file to be as
91 large as the disk. You can assume that the disk will not be larger
92 than 8 MB. In the basic file system, each file is limited to a file
93 size of just under 64 kB. Each file has a header called an index node
94 or @dfn{inode} (represented by @struct{inode}) that is a table of
95 direct pointers to the disk blocks for that file. Since the inode is
96 stored in one disk sector, the maximum size of a file is limited by
97 the number of pointers that will fit in one disk sector. Increasing
98 the limit to 8 MB will require you to implement doubly-indirect
101 @node Problem 4-2 File Growth
102 @section Problem 4-2: File Growth
104 Implement extensible files. In the basic file system, the file size
105 is specified when the file is created. One advantage of this is that
106 the inode data structure, once created, never changes. In UNIX and
107 most other file systems, a file is initially created with size 0 and
108 is then expanded every time a write is made off the end of the file.
109 Modify the file system to allow this. As one test case, allow the
110 root directory file to expand beyond its current limit of ten files.
111 Make sure that concurrent accesses to the inode remain properly
114 @node Problem 4-3 Subdirectories
115 @section Problem 4-3: Subdirectories
117 Implement a hierarchical name space. In the basic file system, all
118 files live in a single directory. Modify this to allow directories to
119 point to either files or other directories. To do this, you will need
120 to implement routines that parse path names into a sequence of
121 directories, as well as routines that change the current working
122 directory and that list the contents of the current directory. For
123 performance, allow concurrent updates to different directories, but
124 use mutual exclusion to ensure that updates to the same directory are
125 performed atomically (for example, to ensure that a file is deleted
128 Make sure that directories can expand beyond their original size just
129 as any other file can.
131 Each process has its own current directory. When one process starts
132 another with the @code{exec} system call, the child process inherits its
133 parent's current directory. After that, the two processes' current
134 directories are independent, so that either changing its own current
135 directory has no effect on the other.
137 Update the existing system calls so that, anywhere a file name is
138 provided by the caller, an absolute or relative path name may used.
139 Also, implement the following new system calls:
143 @itemx bool chdir (const char *@var{dir})
144 Attempts to change the current working directory of the process to
145 @var{dir}, which may be either relative or absolute. Returns true if
146 successful, false on failure.
149 @itemx bool mkdir (const char *dir)
150 Attempts to create the directory named @var{dir}, which may be either
151 relative or absolute. Returns true if successful, false on failure.
154 @itemx void lsdir (void)
155 Prints a list of files in the current directory to @code{stdout}, one
159 Also write the @command{ls} and @command{mkdir} user programs. This
160 is straightforward once the above syscalls are implemented. In Unix,
161 these are programs rather than built-in shell commands, but
162 @command{cd} is a shell command. (Why?)
164 @node Problem 4-4 Buffer Cache
165 @section Problem 4-4: Buffer Cache
167 Modify the file system to keep a cache of file blocks. When a request
168 is made to read or write a block, check to see if it is stored in the
169 cache, and if so, fetch it immediately from the cache without going to
170 disk. (Otherwise, fetch the block from disk into cache, evicting an
171 older entry if necessary.) You are limited to a cache no greater than
172 64 sectors in size. Be sure to choose an intelligent cache
173 replacement algorithm. Experiment to see what combination of accessed,
174 dirty, and other information results in the best performance, as
175 measured by the number of disk accesses. (For example, metadata is
176 generally more valuable to cache than data.) Document your
177 replacement algorithm in your design document.
179 The provided file system code uses a ``bounce buffer'' in @struct{file}
180 to translate the disk's sector-by-sector interface into the system call
181 interface's byte-by-byte interface. It needs per-file buffers because,
182 without them, there's no other good place to put sector
183 data.@footnote{The stack is not a good place because large objects
184 should not be allocated on the stack. A 512-byte sector is pushing the
185 limit there.} As part of implementing the buffer cache, you should get
186 rid of these bounce buffers. Instead, copy data into and out of sectors
187 in the buffer cache directly. You will probably need some
188 synchronization to prevent sectors from being evicted from the cache
189 while you are using them.
191 In addition to the basic file caching scheme, your implementation
192 should also include the following enhancements:
196 Instead of always immediately writing modified data to disk, dirty
197 blocks can be kept in the cache and written out sometime later. Your
198 buffer cache should write behind whenever a block is evicted from the
202 Your buffer cache should automatically fetch the next block of a file
203 into the cache when one block of a file is read, in case that block is
207 For each of these three optimizations, design a file I/O workload that
208 is likely to benefit from the enhancement, explain why you expect it
209 to perform better than on the original file system implementation, and
210 demonstrate the performance improvement.
212 Note that write-behind makes your filesystem more fragile in the face
213 of crashes. Therefore, you should
214 periodically write all cached blocks to disk. If you have
215 @func{timer_sleep} from the first project working, this is an
216 excellent application for it.
218 Likewise, read-ahead is only really useful when done asynchronously.
219 That is, if a process wants disk block 1 from the file, it needs to
220 block until disk block 1 is read in, but once that read is complete,
221 control should return to the process immediately while the read
222 request for disk block 2 is handled asynchronously. In other words,
223 the process will block to wait for disk block 1, but should not block
224 waiting for disk block 2.
226 When you're implementing this, please make sure you have a scheme for
227 making any read-ahead and write-behind threads halt when Pintos is
228 ``done'' (when the user program has completed, etc), so that Pintos
229 will halt normally and the disk contents will be consistent.
231 @node File System Design Document Requirements
232 @section Design Document Requirements
234 As always, submit a design document file summarizing your design. Be
235 sure to cover the following points:
239 How did you structure your inodes? How many blocks did you access
240 directly, via single-indirection, and/or via double-indirection? Why?
243 How did you structure your buffer cache? How did you perform a lookup
244 in the cache? How did you choose elements to evict from the cache?
247 How and when did you flush the cache?
250 @node File System FAQ
255 @b{What extra credit opportunities are available for this assignment?}
259 We'll give out extra credit to groups that implement Unix-style
260 support for @file{.} and @file{..} in relative paths in their projects.
263 We'll give some extra credit if you submit with VM enabled. If you do
264 this, make sure you show us that you can run multiple programs
265 concurrently. A particularly good demonstration is running
266 @file{capitalize} (with a reduced words file that fits comfortably on
267 your disk, of course). So submit a file system disk that contains a
268 VM-heavy program like @file{capitalize}, so we can try it out. And also
269 include the results in your test case file.
271 We feel that you will be much more satisfied with your cs140 ``final
272 product'' if you can get your VM working with your file system. It's
273 also a great stress test for your FS, but obviously you have to be
274 pretty confident with your VM if you're going to submit this extra
275 credit, since you'll still lose points for failing FS-related tests,
276 even if the problem is in your VM code.
279 A point of extra credit can be assigned if a user can recursively
280 remove directories from the shell command prompt. Note that the
281 typical semantic is to just fail if a directory is not empty.
284 Make sure that you discuss any extra credit in your @file{README}
285 file. We're likely to miss it if it gets buried in your design
289 @b{What exec modes for running Pintos do I absolutely need to
292 You also need to support the @option{-f}, @option{-ci}, @option{-co},
293 and @option{-ex} flags individually, and you need to handle them when
294 they're combined, like this: @samp{pintos -f -ci shell 12345 -ex
295 "shell"}. Thus, you should be able to treat the above as equivalent to:
299 pintos -ci shell 12345
303 If you don't change the filesystem interface, then this should already
304 be implemented properly in @file{threads/init.c} and
305 @file{filesys/fsutil.c}.
307 You must also implement the @option{-q} option and make sure that data
308 gets flushed out to disk properly when it is used.
311 @b{Will you test our file system with a different @code{DISK_SECTOR_SIZE}?}
313 No, @code{DISK_SECTOR_SIZE} is fixed at 512. This is a fixed property
314 of IDE disk hardware.
317 @b{Will the @struct{inode} take up space on the disk too?}
319 Yes. Anything stored in @struct{inode} takes up space on disk,
320 so you must include this in your calculation of how many entires will
321 fit in a single disk sector.
324 @b{What's the directory separator character?}
326 Forward slash (@samp{/}).
330 * Problem 4-2 File Growth FAQ::
331 * Problem 4-3 Subdirectory FAQ::
332 * Problem 4-4 Buffer Cache FAQ::
335 @node Problem 4-2 File Growth FAQ
336 @subsection Problem 4-2: File Growth FAQ
340 @b{What is the largest file size that we are supposed to support?}
342 The disk we create will be 8 MB or smaller. However, individual files
343 will have to be smaller than the disk to accommodate the metadata.
344 You'll need to consider this when deciding your @struct{inode}
348 @node Problem 4-3 Subdirectory FAQ
349 @subsection Problem 4-3: Subdirectory FAQ
353 @b{What's the answer to the question in the spec about why
354 @command{ls} and @command{mkdir} are user programs, while @command{cd}
357 Each process maintains its own current working directory, so it's much
358 easier to change the current working directory of the shell process if
359 @command{cd} is implemented as a shell command rather than as another
360 user process. In fact, Unix-like systems don't provide any way for
361 one process to change another process's current working directory.
364 @b{When the spec states that directories should be able to grow beyond
365 ten files, does this mean that there can still be a set maximum number
366 of files per directory that is greater than ten, or should directories
367 now support unlimited growth (bounded by the maximum supported file
370 We're looking for directories that can support arbitrarily large
371 numbers of files. Now that directories can grow, we want you to
372 remove the concept of a preset maximum file limit.
375 @b{When should the @code{lsdir} system call return?}
377 The @code{lsdir} system call should not return until after the
378 directory has been printed. Here's a code fragment, and the desired
382 printf ("Start of directory\n");
384 printf ("End of directory\n");
387 This code should create the following output:
391 ... directory contents ...
396 @b{Do we have to implement both absolute and relative pathnames?}
398 Yes. Implementing @file{.} and @file{..} is extra credit, though.
401 @b{Should @func{remove} also be able to remove directories?}
403 Yes. The @code{remove} system call should handle removal of both
404 regular files and directories. You may assume that directories can
405 only be deleted if they are empty, as in Unix.
408 @node Problem 4-4 Buffer Cache FAQ
409 @subsection Problem 4-4: Buffer Cache FAQ
413 @b{We're limited to a 64-block cache, but can we also keep an
414 @struct{inode_disk} inside @struct{file}, the way the provided code
417 The goal of the 64-block limit is to bound the amount of cached file
418 system data. If you keep a block of disk data anywhere in kernel
419 memory, whether it's file data or metadata, then you have to count it
420 against the 64-block limit. The same rule applies to anything that's
421 ``similar'' to a block of disk data, such as a @struct{inode_disk}
422 without the @code{length} or @code{sector_cnt} members.
424 That means you'll have to change the way the file implementation
425 accesses its corresponding inode right now, since it currently just
426 creates a new @struct{inode} containing an @struct{inode_disk} in
427 @func{inode_open} and reads the corresponding sector in from disk when
430 There are two reasons for not storing inode data in @struct{inode}.
431 First, keeping extra copies of inodes would be cheating the 64-block
432 limitation that we place on your cache. Second, if two processes have
433 the same file open, you will create a huge synchronization headache for
434 yourself if each @struct{inode} has its own copy of the on-disk inode.
436 You can store pointers to inode data in @struct{inode_disk}, if you
437 want, and you can store some other small amount of information to help
438 you find the inode when you need it. Similarly, if you want to store
439 one block of data plus some small amount of metadata for each of your 64
440 cache entries, that's fine.
442 If you look at @func{file_read_at}, it uses the inode directly without
443 having first read in that sector from wherever it was in the storage
444 hierarchy. This will no longer work. You will need to change
445 @code{file_read_at} (and similar functions) so that it reads the inode
446 from the storage hierarchy before using it.