Static Compiler Hot Code Segmentation for Cross-Platform Compatibility

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for optimizing applications to utilize newer processor capabilities often compromise compatibility with older processors, leading to underutilization of advanced features and inefficiencies, such as high overhead in dynamic compilation and large file sizes in fat binaries.

Innovation Solution

A static compiler system that profiles application code to identify 'hot' regions, generates multiple optimized versions for different platform models, and includes navigation instructions to select the appropriate versions at load time based on the target machine's capabilities, enabling efficient exploitation of new capabilities while maintaining backward compatibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If application code is optimized to use newer processor capabilities, then execution performance on new platforms is improved, but compatibility with older processors deteriorates

Engineering Contradiction:

Improveexecution performanceVSAvoidplatform compatibility

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments the application code into multiple versions for different processor platforms. Each hot code region is compiled separately for specific processor capabilities, allowing the system to select the appropriate version at runtime based on the target platform, thus achieving both optimization for new processors and compatibility with older ones

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces dynamic code selection at runtime through navigation instructions that redirect execution to platform-appropriate code versions. The system dynamically determines which code version to execute based on the target processor's capabilities, enabling adaptive performance optimization without sacrificing compatibility

Inventive Principle:

Principle #15Dynamics

2Productivity

If dynamic compilation is used to generate code at runtime, then platform-specific optimization is achieved, but system overhead increases

Engineering Contradiction:

Improveplatform optimizationVSAvoidcompilation overhead

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent performs code compilation in advance during the build process, creating multiple pre-compiled versions for different processor platforms. This eliminates runtime compilation overhead by preparing all necessary code versions beforehand, allowing fast runtime selection without the cost of on-the-fly compilation

Inventive Principle:

Principle #10Preliminary action

3Productivity

If platform-specific libraries are used, then computation acceleration is achieved, but the range of accelerated computations is limited

Engineering Contradiction:

Improvecomputation accelerationVSAvoidcomputation coverage

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal solution that works across multiple processor platforms by compiling the same application code for different target architectures. Instead of requiring separate libraries for different computation types, the system provides a single binary that universally supports multiple platforms through runtime code selection

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Adaptability or versatility

If entire application code is compiled for multiple platforms and stitched together, then compatibility among distinct ISAs is achieved, but binary size increases significantly

Engineering Contradiction:

Improvecross-ISA compatibilityVSAvoidbinary size

Core Design Contradiction:

Adaptability or versatilityVSWeight of stationary object

Solution Approach 1:

The patent segments the binary into modular code regions, each optimized for specific processor capabilities. Instead of including all possible platform variations throughout the entire binary, only the necessary code regions are segmented and included, reducing overall binary size while maintaining cross-platform compatibility

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by optimizing only the specific code regions that benefit from platform-specific capabilities. Non-critical code regions remain unchanged and platform-agnostic, reducing the need for multiple versions and thereby reducing binary size while still achieving compatibility

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS9032379B2Platform specific optimizations in static compilers

Publication Date: 2015.05.12 ORACLE INT CORP

AI summary

Embodiments include systems and methods for generating an application code binary that exploits new platform-specific capabilities, while maintaining backward compatibility with other older platforms. For example, application code is profiled to determine which code regions are main contributors to the runtime execution of the application. For each hot code region, a determination is made as to whether multiple versions of the hot code region should be produced for different target platform models. Each hot code region can be analyzed to determine if benefits can be achieved by exploiting platform-specific capabilities corresponding to each of N platform models, which can result in between one and N versions of that particular hot code region. Navigation instructions are generated as part of the application code binary to permit a target machine to select appropriate versions of the hot code sections at load time, according to the target machine's capabilities.