CS330: Operating Systems
Files
The file system
End-user wants see a
nice tree view. Let me
USER
enable it through a
Storage devices
simple system call
/ APIs.
Hard disk
drive
etc bin sbin home lib OS SSD
code file.txt
Others
The file system
USER Storage devices
OS
/ Hard disk
drive
File system
etc bin sbin home lib SSD
layer
code file.txt
Others
- File system is an important OS subsystem
- Provides abstractions like files and directories
- Hides the complexity of underlying storage devices
File system interfacing
Input/Output Library - Processes identify files through a
(fopen, fclose, fread, fprintf …)
file handle a.k.a. file descriptors
- In UNIX, the POSIX file API is
used to access files, devices,
System call API
(open, close, read, write …) sockets etc.
- What is the mapping between
library functions and system
Files Devices Sockets
calls?
open: getting a handle
int open (char *path, int flags, mode_t mode)
open: getting a handle
int open (char *path, int flags, mode_t mode)
- Access mode specified in flags : O_RDONLY, O_RDWR, O_WRONLY
- Access permissions check performed by the OS
- On success, a file descriptor (integer) is returned
- If flags contain O_CREAT, mode specifies the file creation mode
- Refer man page (“man 2 open")
Process view of file
PCB (P1)
P1
fd1 =open(“file1”) 0 1 2 3 file 1
Inode 1
P2 PCB (P2) file 1
fd1 = open(“file1”)
fd2 = open(“file2”) 0 1 2 3 4 file 2 Inode 2
- Per-process file descriptor table with pointer to a “file” object
- file object → inode is many-to-one
Process view of file
PCB (P1)
P1
fd1 =open(“file1”) 0 1 2 3 file 1
- What do file descriptors 0, 1 and 2 represent? Inode 1
- What P2 happens to the FD PCB
table
(P2)and the filefile
objects across fork( )?
1
fd1 = open(“file1”)
What happens in exec(
fd2 =- open(“file2”) )?
0 1 2 3 4 file 2 Inode 2
- Can multiple FDs point to the same file object?
- Per-process file descriptor table with pointer to a “file” object
Read and Write
ssize_t read (int fd, void *buf, size_t count);
- fd → file handle
- buf → user buffer as read destination
- count → #of bytes to read
- read ( ) returns #of bytes actually read, can be smaller than count
Read and Write
ssize_t read (int fd, void *buf, size_t count);
- fd → file handle
- buf → user buffer as read destination
- count → #of bytes to read
- read ( ) returns #of bytes actually read, can be smaller than count
ssize_t write (int fd, void *buf, size_t count);
- Similar to read
Process view of file
PCB (P1)
P1
fd1 =open(“file1”) 0 1 2 3 file 1
- What do file descriptors 0, 1 and 2 represent? Inode 1
- 0→ P2 STDIN, 1 → STDOUT PCBand
(P2) 2 → STDERR file 1
fd1 = open(“file1”)
- fd2What happens to the FD0 table
= open(“file2”) and the file objects across fork( )?
1 2 3 4 file 2 Inode 2
- What happens in exec( )?
-- Can multiplefile
Per-process FDsdescriptor
point to the same
table withfilepointer
object?to a “file” object
lseek
off_t lseek(int fd, off_t offset, int whence);
- fd → file handle
- offset → target offset
- whence → SEEK_SET, SEEK_CUR, SEEK_END
- On success, returns offset from the starting of the file
lseek
off_t lseek(int fd, off_t offset, int whence);
- fd → file handle
- offset → target offset
- whence → SEEK_SET, SEEK_CUR, SEEK_END
- On success, returns offset from the starting of the file
- Examples
- lseek(fd, SEEK_CUR, 100) → forwards the file position by 100 bytes
- lseek(fd, SEEK_END, 0) → file pos at EOF, returns the file size
- lseek(fd, SEEK_SET, 0) → file pos at beginning of file
File information (stat, fstat)
int stat(const char *path, struct stat *sbuf);
- Returns the information about file/dir in the argument path
- The information is filled up in structure called stat
File information (stat, fstat)
int stat(const char *path, struct stat *sbuf);
- Returns the information about file/dir in the argument path
- The information is filled up in structure called stat
struct stat sbuf;
stat(“/home/user/tmp.txt”, &sbuf);
printf(“inode = %d size = %ld\n”, sbuf.st_ino, sbuf.st_size);
- Other useful fields in struct stat : st_uid, st_mode (Refer stat man page)
Process view of file
PCB (P1)
P1
- What do file descriptors00, 11 and
fd1 =open(“file1”) 2 32 represent?
file 1
Inode 1
- 0 → STDIN, 1 → STDOUT and 2 → STDERR
- What P2 happens to the FD PCB table
(P2)and the filefile
objects
1 across fork( )?
fd1 = open(“file1”)
What happens in exec(
fd2 =- open(“file2”) 0 1 )?2 3 4 file 2 Inode 2
- The FD table is copied across fork( ) ⇒ File objects are shared
-- On exec, openfilefiles
Per-process remain shared
descriptor by default
table with pointer to a “file” object
- Can multiple FDs point to the same file object?
Duplicate file handles (dup and dup2)
int dup(int oldfd);
- The dup() system call creates a “copy” of the file descriptor oldfd
- Returns the lowest-numbered unused descriptor as the new descriptor
- The old and new file descriptors represent the same file
Duplicate file handles (dup and dup2)
int fd, dupfd;
fd = open(“tmp.txt”);
close(1);
dupfd = dup(fd); //What will be the value of dupfd?
printf(“Hello world\n”); // Where will be the output?
Duplicate file handles (dup and dup2)
int fd, dupfd;
fd = open(“tmp.txt”);
close(1);
dupfd = dup(fd); //What will be the value of dupfd?
printf(“Hello world\n”); // Where will be the output?
- Value of dupfd = 1 (assuming STDIN is open)
- “Hello world” will be written to tmp.txt file
Duplicate file handles (dup and dup2)
int dup2(int oldfd, int newfd);
- Close newfd before duping the file descriptor oldfd
- dup2 (fd, 1) equivalent to
- close(1);
- dup(fd);
Duplicate file handles (dup and dup2)
Before dup( )
- Lowest numbered
PCB (P1)
P1 unused fd (i.e., 1) is used
fd1 =open(“file1”) file 1
(Assume STDOUT is
0 1 2 3
closed before)
After dup( )
- Duplicate descriptors
+ dup(fd1) PCB (P1) share the same file state
0 1 2 3 file 1 - Closing one file
descriptor does not
close the file
Use of dup: shell redirection
- Example: ls > tmp.txt
- How implemented?
Use of dup: shell redirection
- Example: ls > tmp.txt
- How implemented?
fd = open (“tmp.txt”)
close( 1); close(2); // close STDOUT and STDERR
dup(fd); dup(fd) // 1→ fd, 2 → fd
exec(ls )
Process view of file
PCB (P1)
P1
- What do file descriptors00, 11 and
fd1 =open(“file1”) 2 32 represent?
file 1
Inode 1
- 0 → STDIN, 1 → STDOUT and 2 → STDERR
- What P2 happens to the FD PCB table
(P2)and the filefile
objects
1 across fork( )?
fd1 = open(“file1”)
What happens in exec(
fd2 =- open(“file2”) 0 1 )?2 3 4 file 2 Inode 2
- The FD table is copied across fork( ) ⇒ File objects are shared
-- On exec, openfilefiles
Per-process remain shared
descriptor by default
table with pointer to a “file” object
- Can multiple FDs point to the same file object?
- Yes, duped FDs share the same file object (within a process)
UNIX pipe( ) system call
PCB (P1)
- pipe( ) takes array of two
P1
pipe(fd[2]) 0 1 2 fd[0] fd[1]
FDs as input
- fd[0] is the read end of
IN (Read)
the pipe
OUT (Write)
- fd[1] is the write end of
Pipe (FIFO Queue) the pipe
- Implemented as a FIFO
queue in OS
UNIX pipe( ) with fork( )
PCB (Parent)
- fork( ) duplicates the file
Parent
pipe(fd[2]) 0 1 2 fd[0] fd[1]
descriptors
- At this point, both the
IN (Read)
parent and the child
OUT (Write)
processes can read/write to
Pipe (FIFO Queue) the pipe
0 1 2 fd[0] fd[1]
PCB (Child)
UNIX pipe( ) with fork( )
PCB (Parent)
- fork( ) duplicates the file
Parent
pipe(fd[2]) 0 1 2 fd[0] fd[1]
descriptors
- close( ) one end of the pipe,
IN (Read)
both in child and parent
OUT (Write)
- Result
Pipe (FIFO Queue) - A queue between
parent and child
0 1 2 fd[0] fd[1]
PCB (Child)
Shell piping : ls | wc -l
PCB (Parent)
- pipe( ) followed by fork( )
Parent
pipe(fd[2]) 0 1 2 fd[0] fd[1]
- Parent: exec( “ls”) after
making STDOUT → out fd
of the pipe (using dup)
OUT (Write)
IN (Read) Pipe (FIFO Queue)
0 1 2 fd[0] fd[1]
PCB (Child)
Shell piping : ls | wc -l
PCB (Parent)
- pipe( ) followed by fork( )
Parent
pipe(fd[2]) 0 1 2 fd[0] fd[1]
- Parent: exec( “ls”) after
making STDOUT → out fd
of the pipe (using dup)
OUT (Write)
- Child: exec(“wc” ) after
IN (Read) Pipe (FIFO Queue) closing STDIN and duping
in fd of pipe
- Result: input of “wc” is
0 1 2 fd[0] fd[1]
connected to output of “ls”
PCB (Child)
