Next | Prev | Top | Contents | Index

Chapter 4: Using the Performance Tools

This chapter explains how to use performance tools to reduce the execution time of your programs. For information about the compiler optimization facilities, see Chapter 5, "Optimizing Program Performance."

This chapter covers the following topics:

"Overview of Profiling"explains how profiling can help you to analyze your data.
"Profiling With prof" describes how to run the profiler, prof and lists its options.
"pc Sampling" explains how to use prof to obtain program counter (pc) sampling.
"Basic Block Counting" covers how to use pixie to perform basic block counting.
"Inclusive Basic Block Counting" explains how to use prof to propagate basic block counts through procedures.
"Profiling Multiprocessed Executables" describes how to profile executables that use sproc and sprocsp system calls.
"Rearranging Procedures With cord" explains how to rearrange procedures to reduce paging and achieve better instruction cache mapping.

Although it may be possible to obtain short-term speed increases by relying on unsupported or undocumented quirks of the compiler system, it's a bad idea to do so. Any such "features" may break in future releases of the compiler system. The best way to produce efficient code that can be trusted to remain efficient is to follow good programming practices; in particular, choose good algorithms and leave the details to the compiler.

The techniques described in this manual comprise only a part of performance tuning. Other areas that you can tune, but are outside the scope of this document, include graphics, I/O, the kernel, system parameters, memory, and real-time system calls.

Overview of Profiling
Profiling With prof
pc Sampling
Basic Block Counting
Inclusive Basic Block Counting
Profiling Multiprocessed Executables
Rearranging Procedures With cord

Next | Prev | Top | Contents | Index