-mtune (-qtune)

Pragma equivalent

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.

Syntax


                 .-power8-.   
                 +-pwr8---+   
>>- -m--tune--=--+-pwr9---+------------------------------------><
                 '-power9-'


                 .-balanced-.                        
>>- -q--tune--=--+-pwr8-----+--+-----------------+-------------><
                 +-pwr9-----+  |    .-st-------. |   
                 '-auto-----'  '-:--+-balanced-+-'   
                                    +-smt2-----+     
                                    +-smt4-----+     
                                    '-smt8-----'

Defaults

-mtune=pwr8 or -mtune=power8

-qtune=balanced:balanced when no valid -qarch setting is in effect. Otherwise, the default depends on the effective -qarch setting. You can find details in Acceptable -qarch and -qtune combinations.

Parameters for CPU suboptions

The following 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. You can specify the auto suboption with -qtune only.
balanced: Optimizations are tuned across a selected range of recent hardware. You can specify the balanced suboption with -qtune only.
power8: Optimizations are tuned for the POWER8® hardware platform. You can specify this suboption with -mtune only.
power9: Optimizations are tuned to utilize the POWER9™ technology. You can specify this suboption with -mtune only.
pwr8: Optimizations are tuned for the POWER8 hardware platform.
pwr9: Optimizations are tuned to utilize the POWER9 technology.

Parameters for SMT suboptions

The following simultaneous multithreading (SMT) suboptions allow you to optionally specify an execution mode for the compiler to target for best performance. You can specify these SMT suboptions with -qtune only.

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

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

Although changing the -mtune or -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.

You can find the acceptable combinations of -qarch and -qtune in the following table.

Table 1. Acceptable -qarch and -qtune combinations
-qarch option	Default -qtune setting	Available -qtune CPU settings	Available -qtune SMT settings
pwr8	pwr8:st	auto \| pwr8 \| pwr9 \| balanced	balanced \| st \| smt2 \| smt4 \| smt8
pwr9	pwr9:st	auto \| pwr9 \| balanced	balanced \| st \| smt2 \| smt4 \| smt8

Predefined macros