Tutorial Week 3

Questions

    System Call Interface

  1. The following segment of code is similar (but much simpler) to the main task that the daemon inetd performs. It accepts connections on a socket and forks a process to handle the connection.

    This is not guaranteed to be compilable. Use the man command if you want to investigate what all the system calls are doing.

    0001 xxx(int socket){
    0002 
    0003        while ((fd = accept(socket, NULL, NULL)) >= 0) {
    0004                switch((pid = fork())) {
    0005                case -1:
    0006                        syslog(LOG_WARN, "%s cannot create process: %s",
    0007				progname, sys_error(errno));
    0008                        continue;
    0009                case 0:         
    0010                        close(0);
    0011                        close(1);
    0012                        dup(fd);
    0013                        dup(fd);
    0014                        execl("/usr/sbin/handle_connection", 
    0015				"handle_connection", NULL);
    0016                        syslog(LOG_WARN, "%s cannot exec handle_connection\ 
    0017				helper : %s", progname, sys_error(errno));
    0018                        _exit(0);
    0019                default: 
    0020                        waitpid(pid, &status, 0); 
    0021                        if (WIFEXITED(status) && WIFEXITSTATUS(status) == 0)
    0022                                continue;
    0023                        syslog(LOG_WARN, "handle_connection failed:\
    0024				exit status +%d\n", status);
    0025                }
    0026        }
    0027 }
    
    1. Identify which lines of code are executed by the parent process.
    2. Identify which lines of code are invoked by the child process.
    3. Under what circumstances does the child terminate?

  2. R3000 and assembly

  3. What is a branch delay?

  4. The goal of this question is to have you reverse engineer some of the C compiler function calling convention (instead of reading it from a manual). The following code contains 6 functions that take 1 to 6 integer arguments. Each function sums its arguments and returns the sum as a the result.

    #include <stdio.h> /* function protoypes, would normally be in header files */ int arg1(int a); int arg2(int a, int b); int arg3(int a, int b, int c); int arg4(int a, int b, int c, int d); int arg5(int a, int b, int c, int d, int e ); int arg6(int a, int b, int c, int d, int e, int f); /* implementations */ int arg1(int a) { return a; } int arg2(int a, int b) { return a + b; } int arg3(int a, int b, int c) { return a + b + c; } int arg4(int a, int b, int c, int d) { return a + b + c + d; } int arg5(int a, int b, int c, int d, int e ) { return a + b + c + d + e; } int arg6(int a, int b, int c, int d, int e, int f) { return a + b + c + d + e + f; } /* do nothing main, so we can compile it */ int main() { }

    The following code is the disassembled code that is generated by the C compiler (with certain optimisations turned of for the sake of clarity).

    004000f0 <arg1>: 4000f0: 03e00008 jr ra 4000f4: 00801021 move v0,a0 004000f8 <arg2>: 4000f8: 03e00008 jr ra 4000fc: 00851021 addu v0,a0,a1 00400100 <arg3>: 400100: 00851021 addu v0,a0,a1 400104: 03e00008 jr ra 400108: 00461021 addu v0,v0,a2 0040010c <arg4>: 40010c: 00852021 addu a0,a0,a1 400110: 00861021 addu v0,a0,a2 400114: 03e00008 jr ra 400118: 00471021 addu v0,v0,a3 0040011c <arg5>: 40011c: 00852021 addu a0,a0,a1 400120: 00863021 addu a2,a0,a2 400124: 00c73821 addu a3,a2,a3 400128: 8fa20010 lw v0,16(sp) 40012c: 03e00008 jr ra 400130: 00e21021 addu v0,a3,v0 00400134 <arg6>: 400134: 00852021 addu a0,a0,a1 400138: 00863021 addu a2,a0,a2 40013c: 00c73821 addu a3,a2,a3 400140: 8fa20010 lw v0,16(sp) 400144: 00000000 nop 400148: 00e22021 addu a0,a3,v0 40014c: 8fa20014 lw v0,20(sp) 400150: 03e00008 jr ra 400154: 00821021 addu v0,a0,v0 00400158 <main>: 400158: 03e00008 jr ra 40015c: 00001021 move v0,zero
    1. arg1 (and functions in general) returns its return value in what register?
    2. Why is there no stack references in arg2?
    3. What does jr ra do?
    4. Which register contains the first argument to the function?
    5. Why is the move instruction in arg1 after the jr instruction.
    6. Why does arg5 and arg6 reference the stack?

  5. What is the EPC register? What is it used for?

  6. What happens to the KUc and IEc bits in the STATUS register when an exception occurs? Why? How are they restored?

  7. What is the value of ExcCode in the Cause register immediately after a system call exception occurs?

  8. The following code provides an example to illustrate stack management by the C compiler. Firstly, examine the C code in the provided example to understand how the recursive function works.

    #include <stdio.h> #include <unistd.h> char teststr[] = "\nThe quick brown fox jumps of the lazy dog.\n"; void reverse_print(char *s) { if (*s != '\0') { reverse_print(s+1); write(STDOUT_FILENO,s,1); } } int main() { reverse_print(teststr); }

    The following code is the disassembled code that is generated by the C compiler (with certain optimisations turned off for the sake of clarity).

    1. Describe what each line in the code is doing.
    2. What is the maximum depth the stack can grow to when this function is called?
    004000f0 <reverse_print>: 4000f0: 27bdffe8 addiu sp,sp,-24 4000f4: afbf0014 sw ra,20(sp) 4000f8: afb00010 sw s0,16(sp) 4000fc: 80820000 lb v0,0(a0) 400100: 00000000 nop 400104: 10400007 beqz v0,400124 <reverse_print+0x34> 400108: 00808021 move s0,a0 40010c: 0c10003c jal 4000f0 <reverse_print> 400110: 24840001 addiu a0,a0,1 400114: 24040001 li a0,1 400118: 02002821 move a1,s0 40011c: 0c1000af jal 4002bc <write> 400120: 24060001 li a2,1 400124: 8fbf0014 lw ra,20(sp) 400128: 8fb00010 lw s0,16(sp) 40012c: 03e00008 jr ra 400130: 27bd0018 addiu sp,sp,24

  9. Why is recursion or large arrays of local variables avoided by kernel programmers?

  10. Why must kernel programmers be especially careful when implementing system calls?

  11. The following questions are focused on the case study of the system call convention used by OS/161 on the MIPS R3000 from the lecture slides.
    1. How does the 'C' function calling convention relate to the system call interface between the application and the kernel?
    2. What does the most work to preserve the compiler calling convention, the system call wrapper, or the OS/161 kernel.
    3. At minimum, what additional information is required beyond that passed to the system-call wrapper function?

  12. In the example given in lectures, the library function read invoked the read system call. Is it essential that both have the same name? If not, which name is important?

  13. To a programmer, a system call looks like any other call to a library function. Is it important that a programmer know which library function result in system calls? Under what circumstances and why?
  14. Processes and Threads

  15. In the three-state process model, what do each of the three states signify? What transitions are possible between each of the states, and what causes a process (or thread) to undertake such a transition?

  16. Given N threads in a uniprocessor system. How many threads can be running at the same point in time? How many threads can be ready at the same time? How many threads can be blocked at the same time?

  17. Compare reading a file using a single-threaded file server and a multithreaded file server. Within the file server, it takes 15 msec to get a request for work and do all the necessary processing, assuming the required block is in the main memory disk block cache. A disk operation is required for one third of the requests, which takes an additional 75 msec during which the thread sleeps. How many requests/sec can a server handled if it is single threaded? If it is multithreaded?