Linking code for an enhanced application binary interface (ABI) with decode time instruction optimization
A technique of instruction and code sequence, applied in the field of compiled code, which can solve the problem of not knowing the final parsed value, etc.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Publication Date
- 2014-06-11
- Estimated Expiration
- Not applicable · inactive patent
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The present invention relates to computer systems, and more particularly to optimized linking of compiled code for an enhanced Application Binary Interface (ABI) using decode time instructions. Background technique
[0002] In computer software, an Application Binary Interface (ABI) describes the low-level interface between an application program and the operating system, or between an application program and another application. The ABI involves details such as data types, sizes, and alignments; calling rules that govern how function variables are passed and how return values are searched for; system call numbers and how applications should make system calls to the operating system; In the case of an ABI, it refers to the binary format of object files, libraries, and so on. Several ABIs (e.g., the Interactive UnixABI that allows a wide range of programs to run on the various Unix and Linux variants for the Intel x86 architecture) allow access from ...
Examples
Embodiment Construction
[0025] Embodiments of the invention relate to minimizing the impact of table of contents (TOC) overflow on performance and throughput in computing systems. One embodiment includes a compiler designed to generate object code to include specified instruction sequences (eg, sequences that have been inserted into the code to compensate for TOC overflow). The sequence of instructions is adapted to be recognized by hardware so that the microprocessor can optimize the sequence for internal execution. When one of the specified instruction sequences is located, the microprocessor replaces the instruction in the sequence with an internal instruction that can be executed more efficiently, or replaces the instruction in the sequence with a single internal instruction. This processing performed by the microprocessor is referred to herein as Decode Time Instruction Optimization (DTIO).
[0026] DTIO processing is a hardware processing. The compiler and linker described here prepare code s...