malloc: optimizations (#16037)

[reusee]

What type of PR is this?

• API-change
• BUG
• Improvement
• Documentation
• Feature
• Test and CI
• Code Refactoring

Which issue(s) this PR fixes:

issue #15479

What this PR does / why we need it:

malloc: refine cache size calculation

malloc: remove AllocTyped

malloc: remove eviction

malloc: add shard alloc and free statistics

malloc: flush cached objects when idle

[matrix-meow]

@reusee Thanks for your contributions!

Pull Request Review:

Title: malloc: optimizations (#16037)

Body:

• The PR is categorized as an Improvement.
• It addresses issue [Refactoring]: Remove or replace mmap with malloc #15479.
• The changes include refining cache size calculation, removing AllocTyped, removing eviction, adding shard alloc and free statistics, and flushing cached objects when idle.

Changes:

1. Bench_test.go:

   • Removed BenchmarkAllocTyped and BenchmarkParallelAllocTyped functions.
   • This change seems appropriate as these functions are no longer needed.

2. Flush.go:

   • Added a new file flush.go containing a function to flush cached objects when idle.
   • The implementation looks fine and adds a necessary feature.

3. Handle.go:

   • Modified the Free method to use a shard based on the processor ID for better distribution.
   • This change improves the distribution of objects across shards.

4. Malloc.go:

   • Replaced some global variables with a Shard struct to manage allocation and deallocation statistics per shard.
   • Refactored the initialization of shards and pools to improve memory management.
   • Removed AllocTyped function which is no longer used.
   • The changes seem to optimize memory allocation and deallocation processes.

Suggestions for Improvement:

1. Code Refactoring:

   • Consider breaking down the changes into smaller, more focused commits for better readability and easier review.
   • Ensure consistent formatting and naming conventions across the codebase.

2. Error Handling:

   • Add error handling mechanisms where necessary, especially in functions that involve critical operations like memory allocation and deallocation.

3. Documentation:

   • Provide clear and concise comments for new functions, structs, and significant code blocks to enhance code maintainability and readability.

4. Testing:

   • Ensure that the changes are thoroughly tested to validate the performance improvements and prevent regressions.

5. Security Considerations:

   • Review the changes for any potential security vulnerabilities, especially in memory management operations, to prevent exploitation.

6. Performance Optimization:

   • Continuously monitor the performance impact of the optimizations introduced and fine-tune them if needed for better efficiency.

By addressing the suggestions mentioned above, the codebase can be further improved in terms of readability, maintainability, and performance optimization.