TNS/R Native Application Migration Guide

Introduction to Native Mode
TNS/R Native Application Migration Guide523745-001
1-4
Differences Between Accelerated and Native Object
Code
Differences Between Accelerated and Native Object Code
While both accelerated and native object code execute RISC instructions, most native
object code has a significant performance advantage over accelerated object code.
The Accelerator cannot produce RISC instructions for TNS instruction sequences
whose exact meaning cannot be determined until run time. In such cases, a process
makes a transition into TNS code and executes the TNS instructions through millicode
routines. To enable this transition to occur, the accelerated object code maintains TNS
architecture-specific constructs, such as the P and ENV registers. Native object code
consists entirely of RISC instructions; transitions into TNS code do not occur so TNS
architecture-specific constructs are not maintained. Additionally, the Accelerator must
base its RISC instruction sequences, data layout, and code optimizations on object
code. The native compilers base their RISC instructions sequences, data layout, and
code optimizations on source code. For more information on the Accelerator, see the
Accelerator Manual.
Native Development Environment
A development environment comprises the tools used to compile, link, optimize, and
debug a program and the run-time libraries available to a program. The native
development environment includes:
pTAL Compiler on page 1-5
Native C Compiler on page 1-6
Native C++ Compiler on page 1-6
Native COBOL85 Compiler on page 1-7
Native C Run-Time Library on page 1-7
Native Link Editor (nld Utility) on page 1-8
Native Object File Tool (noft Utility) on page 1-8
Inspect and Visual Inspect Symbolic Debuggers and Debug Facility on page 1-9
SQL Compiler on page 1-9
Data Definition Language (DDL) on page 1-10