M5: File System

The SOS file system features a flat directory structure, containing files and the terminal device console. Files have attributes (like size, creation time, permissions) that can be accessed through appropriate systems calls. All I/O primitives provided by SOS are synchronous (blocking).

You will need to maintain file descriptors containing a present read position in the file. This requires some minimal process descriptor block (PCB) data structure. Again, don't worry at this stage about the exact final contents of the PCB, you can extend them later on. You may wish to initially only deal with a single client process, data on which is kept in a single PCB variable. This avoids dealing with process IDs for the time being.

The low-level file system is implemented by the NFS file system on your host machine. You will be provided with an NFS client library for accessing files on your host Linux machine.

Code updates

This milestone comes with some code updates. These code updates provide you with three new libraries, libgt, libtulip, liblwip and libnfs.

You will need update your Makefile to build these extra libraries, and to link them in with your OS. You will also need to update the INCLUDES variable in the Makefile so that the the lwIP headers are available:

   INCLUDES := -I./include -Ilib/lwip/include -Ilib/lwip/include/ipv4

Updated copies of hardware.c and hardware.h are also available.

Network card driver

Before you can use the NFS library you need to set up a network card driver. This requires you to start libgt (the PCI driver) and libtulip (the network card driver). There is a new hardware.c that takes care of the details for you. If you have modified the original hardware.c (provided earlier) you should make sure you patch your changes carefully into the new one. If you are interested in the exact details involved in initialising the network driver you should read the provided code.

IP stack and NFS

In the code you are provided with a sos/network.c file, which provides network_init(). This should be called after the hardware_init() function has executed. Ensure this is called in a thread different from your interrupt handler or else you will encounter deadlock.

Apart from the initialisation and mounting filesystems, the NFS library provides an asynchronous interface, which means when you call the function, you must also provide a callback function which will be called when the transaction is completed. This should make it easy to provide a blocking interface each clients, without blocking the whole system.

The NFS server you will be talking to is your host machine. The filesytem you mount is the /tftpboot.

Design issues

One of the major issues you will need to deal with in this milestone is converting an asynchronous interface (as provided by libnfs) into a synchronous interface (as required by the system call interface).

If you are planning on doing the file system caching, or dynamic file system advanced component you should probably consider them in your design now.

Assessment

You must have some test code to demonstrates that you have implemented the open, close, read, write, getdirent and stat system calls. I would suggest adding a cp command to sosh, this will test both reading and writing a file. A cp command, together with the supplied ls command, should be sufficient to show that your implementation works.

As always you should be able to explain any design decisions you made, specifically, the design of your open file table and how you dealt with the asynchronous/synchronous problem.