Malloc

Project Overview

Malloc is widely considered one of the most challenging system projects. The goal is to rewrite the standard C library memory management functions (malloc, free, calloc, realloc) to create a shared library capable of replacing the system's default allocator.
Grade achieved: 97% (Near-perfect implementation including advanced optimizations).

Technical Deep Dive

This project required a mastery of pointer arithmetic, memory alignment, and system calls. Instead of using sbrk (deprecated), I utilized mmap to request pages directly from the kernel.

Memory Block Management

Memory is managed via a linked list of metadata headers embedded directly within the allocated pages. I implemented block splitting and coalescing (merging adjacent free blocks) to minimize fragmentation and reduce memory footprint.

Performance & Features

Feature	Implementation Detail
Allocation Strategy	Implementation of "Best-Fit" or "First-Fit" algorithms to speed up block finding while keeping memory usage low.
Realloc Optimization	Smart realloc that extends the current block if space allows, avoiding expensive memory copies (`memcpy`) when possible.
Alignment	Strict memory alignment (on long double boundaries) to prevent bus errors and ensure CPU efficiency.
Robustness	Ability to run complex real-world software (like ls, vim, or gcc) by preloading the library (LD_PRELOAD).

Conclusion

This project was a definitive exercise in system optimization. achieving 97% required not just code that works, but code that is fast, efficient, and proofed against concurrency issues.