Week 11.a
CS3650
03/18 2024
https://naizhengtan.github.io/24spring/

1. FS interface
2. Lab4: fs3650
------

1. FS interfaces

  --statfs  - report file system statistics
  --fstat   - get attributes of a file/directory
  --readdir - enumerate entries in a directory
  --read    - read data from a file

  --rename   - rename a file
  --chmod    - change file permissions
  --creat    - create a new (empty) file
  --mkdir    - create new (empty) directory
  --unlink   - remove a file
  --rmdir    - remove a directory
  --write    - write to a file
  --truncate - delete the contents of a file
  --utime    - change access and modification times

  Question:
  What syscalls do the following cmd use?
  ("strace" is used to track syscalls:
    $ strace [CMD] 2>&1 | grep 'statfs\|fstat\|readdir\|read\|rename\|chmod\|creat\|mkdir\|unlink\|rmdir\|write\|truncate\|utime\|open\|close'
  )

  Question: "$ df"?

  Question: "$ cat file"?

  Question: "$ mkdir dir1"?


2. CS3650 File System (lab4)

  [this is an (almost) duplication of the lab4 instructions.]

  - a Unix-like fs
    -- that is read only
    -- with many simplifications
    -- for example, dirs are no deeper than 10

  A. fs3650 disk

    - an abstract disk
      (the overall system format)

      -- a block is 4KB

     +-------+----------+------------------------+
     | super | root dir |     data blocks ...    |
     | block |   inode  |                        |
     +-------+----------+------------------------+
    block 0       1         2          3 ...

    - talk to the disk?
      -- read or write one or multiple blocks

      [see handout]

  B. important data structures

    ** superblock
      [see handout]

    ** file and inode

      [see handout]

      -- metadata
        [or stat, see full file metadata: https://pubs.opengroup.org/onlinepubs/7908799/xsh/sysstat.h.html]

        -- uid, gid: user and group of this file
        -- mode (see below)
        -- ctime: changed time, time of last status (metadata) change
        -- mtime: modified time, time of last data modification
        (fs3650 skips the access time in Unix)
        -- size: file size in bytes

      -- ptrs

        [draw on board]

    ** mode (uint32_t)

     Question: how does fs3650 tell if an inode is a file or a dir?

     [see handout]

      |<-- S_IFMT --->|           |<-- user ->|<- group ->|<- world ->|
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
      | F | D |   |   |   |   |   | R | W | X | R | W | X | R | W | X |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+

      -- R: read
      -- W: write
      -- X: execute

      -- F bit: 0100000 (the 16th bit from right)
      -- D bit: 0040000 (the 15th bit from right)

   ** directory

     -- recall that directory is used to map name to #inode

     [see handout]
     [draw on board as well]

     --Question: how many files can appear in one directory?
       fs3650 only supports using 1 data block (our choice).

     [answer:
      one fs_dirent is 32B (see handout)
      one data block is 4KB.
      fs3650 chooses to use only 1 data block for dir,
      so the max files in one dir is 4KB/32B=128.
     ]

  C. critical functionalities

    ** path walk

       input: a path (a string, char *path)
       output: inode number (int)

       --how? for example, "/a/b/c/file"
       [answer:
         1. split path to tokens: ["a", "b", "c", "file"]
         2. starting from root dir (block 1)
         3. find "a" in "/" (block 1)
         4. get "a"'s inode
         5. get "b" in "/a/"
         6. ...
       ]

    ** get attributes of a file (fs_getattr)

       run "$ ls -l":

       -rwxrw-r--  1 usr grp 5558 Jul 27 2023  llvm.sh
          ^        ^  ^   ^   ^       ^          ^
          |        |  |   |   |       |          |
      permission #link| group |   modified date  |
                     user    size (byte)       file name

      Question: where are these attributes stored in fs?
      [answer: inode]

      how to get these? by fs_getattr

    ** read data from a file (fs_read)

      --how? for example, "read("/a/file", buf, len, offset)" (pseudocode)
      [answer:
        1. path walk to find the file's inode
        2. find offset's block
           (all data blocks compose a linear space)
        3. read len bytes
      ]

    Question: how many block_reads does the
        "read(/a/file", buf, 10, 4096)" consume?

      [answer: 7 blocks
          - root inode
          - root data block
          - "/a" inode
          - "/a" data block
          - "/a/file" inode
          - "/a/file"'s first data block
          - "/a/file"'s second data block
      ]