|
(a) Describe the three objectives of
translation. [3]
To convert a high level language into machine
language (1 mark), to
identify any source code errors (1 mark), and to produce compact and
efficient object code (1 mark). [3 marks]
(b) What are the three tasks involved in lexical analysis? [3]
Changing source code into a form that is
independent of the input device (1 mark), standardising formats (1
mark), and converting reserved words and composite symbols into tokens
(1 mark). [3 marks]
(c) Give two reasons why programs that are converted into machine
language by
translators cannot be executed directly. [2]
One mark should be awarded for each named
reason (up to a maximum of two marks). The reasons are as follows:
- The program may call subroutines that were not converted to machine
language at the same time as the current program and, as such, their
addresses are unknown.
- The program may be a subprogram and may require a calling program
to work.
- The program may not be placed in memory at the starting address
assumed by the translator. [2 marks]
(d) (i) Describe the term compile-and-go loader. [2]
A compile-and-go loader involves translators
running in one part of memory and the translated machine instructions
and data are assembled and placed directly into their assigned memory
locations (1 mark). When the program assembly is completed, the loader
transfers to the starting instruction of the newly translated
program (1 mark). [2 marks]
(ii) List two disadvantages of compile-and-go loaders [2]
One mark should be awarded for each named
disadvantage (up to a maximum of two marks). Examples include the
following:
- The portion of memory occupied by the translator is wasted because
the object code cannot use that much of memory.
- The program must be translated each time it is run.
- It is difficult to handle multiple segments, especially if different
source codes are in different languages.
- It is difficult to produce orderly modular programs.
Other correct answers should also receive credit. [2 marks]
(e) (i) How do absolute loaders differ from compile-and-go loaders?
[1]
Absolute loaders differ from compile-and-go
loaders in that object
code is not stored directly in memory. [1 mark]
(ii) List two disadvantages of absolute loaders. [2]
One mark should be awarded for each named
disadvantage (up to
a maximum of two marks). Examples include the following:
- The programmer must specify to the assembler the address at which
the program is to be loaded.
- If there are many subroutines, the programmer must remember the
address of each and use that absolute address explicitly in other
subroutines to perform subroutine linkage.
- If the program size increases, necessary adjustments must be made
to all subroutines that access it.
Other correct answers should also receive credit. [2 marks]
|