GATE CS 2014 Set 3 Q.27 (CS Q.17): Intermediate Code Generation, Data Flow Analysis and Code Optimization

One of the purposes of using intermediate code in compilers is to

make parsing and semantic analysis simpler.

improve error recovery and error reporting.

increase the chances of reusing the machine-independent code optimizer in other compilers.

improve the register allocation.

Official answer and solution

Solution

Recall the fundamental role of intermediate representation (IR) in decoupling compiler front-ends from target architecture back-ends.

In compiler design, intermediate representations (IR) separate the source-language specific front-end (lexical analysis, parsing, semantic analysis) from the target-machine specific back-end (instruction selection, register allocation, machine code generation).
By using a machine-independent intermediate code:
- Machine-independent optimizations (such as common subexpression elimination, loop invariant code motion, dead code elimination, and constant folding) can be implemented once on the IR.
- This optimizer module can then be reused across different source languages and target architectures without modifying machine-specific code.
Other options are incorrect:
- Parsing and semantic analysis precede intermediate code generation, so IR does not simplify them (option A).
- Error detection and reporting predominantly occur in the front-end (option B).
- Register allocation is an architecture-specific phase performed during target code generation (option D).

Hence, intermediate code allows reusing the machine-independent code optimizer in other compilers (option C).

Final answer

Common trap

Thinking intermediate code simplifies parsing, whereas parsing happens before intermediate code is generated.

The answer is the one in the official GATE answer key. The solution is this site's own explanation.

GATE questions are the work of the organising institutes. This site is independent and is not affiliated with them.