- Minimized execution time and stack size
- Minimized RAM and ROM resources
- Target-optimized code
Efficiency is the key to production-quality code. Efficient code means that a minimum of execution time, RAM resources, ROM resources, and stack size is required to run the code on an embedded processor. TargetLink® meets these high standards for production-quality code generation through a variety of technological innovations.
TargetLink's code optimization capabilities fall into three main groups:
- Traditional optimizations, also used by C compilers
- Interblock optimization to combine the code of several blocks into one C code line
- Code pattern libraries for highly efficient code for complex blocks such as filters
ANSI C Code and Target Optimized Code
TargetLink generates highly efficient, microcontroller-independent ANSI C code that can be used for embedded applications with virtually any microcontroller. Moreover, for many microcontrollers in common use, even greater efficiency is achieved by using Target Optimization Modules to generate microcontroller-specific C code.
Target Optimization Modules
Processor- and compiler-specific language extensions are yet another code optimization technique. These are provided by the Target Optimization Modules, which create compiler-specific instructions and even inline assembly macros from libraries of highly optimized code patterns. This again boosts the efficiency of the code and makes a big difference if resources on the target are scarce.
- Freescale HCS12 (Metrowerks CodeWarrior compilers, Cosmic compilers)
- Freescale MPC5xx (Wind River compilers, Green Hills compilers)
- Freescale MPC55xx (Wind River compilers)
- Infineon C16x (Tasking compilers)
- Infineon TriCore (Tasking compilers)
- Renesas M32R (GAIO compilers)
- Renesas SH-2, SH-2e cores only (Renesas compilers)