-qtune

Category

Optimization and tuning

@PROCESS

None.

Purpose

Tunes instruction selection, scheduling, and other architecture-dependent performance enhancements to run best on a specific hardware architecture. Allows specification of a target SMT mode to direct optimizations for best performance in that mode.

Defaults

-qtune=balanced:balanced when no valid -qarch setting is in effect. Otherwise, the default depends on the effective -qarch setting. See Table 1 for details.

Parameters for CPU suboptions

The following -qtune CPU suboptions allow you to specify a particular architecture for the compiler to target for best performance:

auto

Optimizations are tuned for the platform on which the application is compiled.

balanced

Optimizations are tuned across a selected range of recent hardware.

pwr5

Optimizations are tuned for the POWER5 hardware platforms.

pwr6

Optimizations are tuned for the POWER6® hardware platforms.

pwr7

Optimizations are tuned for the POWER7® or POWER7+™ hardware platforms.

pwr8

Optimizations are tuned for the POWER8™ hardware platforms.

Parameters for SMT suboptions

The following -qtune simultaneous multithreading (SMT) suboptions allow you to optionally specify an execution mode for the compiler to target for best performance:

balanced

Optimizations are tuned for performance across various SMT modes for a selected range of recent hardware.

st

Optimizations are tuned for single-threaded execution.

smt2

Optimizations are tuned for SMT2 execution mode (two threads).

smt4

Optimizations are tuned for SMT4 execution mode (four threads).

smt8

Optimizations are tuned for SMT8 execution mode (eight threads).

Usage

If you want your program to run on more than one architecture, but to be tuned to a particular architecture, you can use a combination of the -qarch and -qtune options. These options are primarily of benefit for floating-point intensive programs.

By arranging (scheduling) the generated machine instructions to take maximum advantage of hardware features such as cache size and pipelining, -qtune can improve performance. It only has an effect when used in combination with options that enable optimization.

A particular SMT suboption is valid if the effective -qarch option supports the specified SMT mode. The acceptable combinations of the -qarch and SMT tune options are listed in Table 1. The compiler ignores any invalid -qarch/-qtune SMT combination.

Although changing the -qtune setting may affect the performance of the resulting executable, it has no effect on whether the executable can be executed correctly on a particular hardware platform.

Acceptable combinations of -qarch and -qtune are shown in the following table.

Related information

• -qarch
• Optimizing your applications