Chapter 6: Processes

Exercise 6-1

The source code is downloaded from mem_segments.c.

To do the exercise:

1. Use make mem_segments to compile and link the code.
2. Use ls -l mem_segments to see the size of the code.

The output should be similar to:

-rwxr-xr-x 1 douglas douglas 18848 Nov  4 11:22 mem_segments

The answer is given on p. 116:

...The main reason for placing global and static variables that are initialized into a separate segment from those that are unitialized is that, when a program is stored on disk, it is not necessary to allocate space for unitialized data. Instead, the executable merely needs to record the location and size required for the uninitialized data segment, and this space is allocated by the praogram loader at run time.

Exercise 6-2

My solution is in ex6_2.c which is a modified version of longjmp.c. The real solution can be found in bad_longjmp.c.

To do the exercise:

1. Use make ex6_2 to compile and link the code.
2. Use ./ex6_2 to execute the solution.

The output should be similar to:

Calling f1() after initial setjmp()
test_jmp(1) finished
We jumped back from f1()
test_jmp(1) finished

In effect, the body of the function is executed twice.

However, the solution (./bad_longjmp) provided does the following:

Entered setJump
Entered setJump2
Exiting setJump2
Exiting setJump
Entered doJump
Exiting setJump2
Entered setJump2
Exiting setJump2
Segmentation fault (core dumped)

Exercise 6-3

My solution is in ex6_3.c. The real solution is in setenv.c.

1. Use make ex6_3 to compile and link the code.
2. Use ./ex6_3 to test the solution.

The expected output is:

FRED=NURK
FRED=NURK
FRED=charlie
FRED=NULL

To run the real solution:

1. Use make setenv to compile and link the code.
2. Use ./setenv to execute the code.

The output should be similar to:

***** Environment before unsetenv(TT)
TT=xxxxx
Total lines from printenv: 58
***** Environment after unsetenv(TT)
Total lines from printenv: 57
***** Environment after setenv() calls
xyz=one
xyz2=222
Total lines from printenv: 59