-qsimd
Category
Pragma equivalent
#pragma nosimd
Purpose
Controls whether the compiler can automatically take advantage of vector instructions for processors that support them.
These instructions can offer higher performance when used with algorithmic-intensive tasks such as multimedia applications.
Defaults
- The optimization level is -O3 or higher.
- -qarch is set to pwr7 or higher.
Usage
The -qsimd=auto option enables automatic generation of vector instructions for processors that support them. When -qsimd=auto is in effect, the compiler converts certain operations that are performed in a loop on successive elements of an array into vector instructions. These instructions calculate several results at one time, which is faster than calculating each result sequentially. These options are useful for applications with significant image processing demands.
The -qsimd=noauto option disables the conversion of loop array operations into vector instructions. To achieve finer control, use -qstrict=ieeefp, -qstrict=operationprecision, and -qstrict=vectorprecision. For details, see -qstrict.
The -qsimd=auto option controls the autosimdization, which was performed by the deprecated -qhot=simd option. If you specify -qhot=simd, the compiler ignores it and does not issue any warning message.
Specifying the deprecated -qenablevmx option has the same effect as specifying -qsimd=auto. The compiler does not issue any warning for this.
- Specifying -qsimd without any suboption is equivalent to -qsimd=auto.
- Specifying -qsimd=auto does not guarantee that autosimdization will occur.
- Using vector instructions to calculate several results at one time might delay or even miss detection of floating-point exceptions on some architectures. If detecting exceptions is important, do not use -qsimd=auto.
Rules
If you enable IPA and specify -qsimd=auto at the IPA compile step, but specify -qsimd=noauto at the IPA link step, the compiler automatically sets -qsimd=auto at the IPA link step. It also sets an appropriate value for -qarch to match the architecture that is specified at the compile time. Similarly, if you enable IPA and specify -qsimd=noauto at the IPA compile step, but specify -qsimd=auto at the IPA link step, the compiler automatically sets -qsimd=auto at the compile step.
Predefined macros
None.
Examples
- xlc -O3 -qsimd
- xlc -O2 -qhot=level=0 -qsimd=auto
- xlc -O2 -qsimd=auto
...
#pragma nosimd
for (i=1; i<1000; i++) {
/* program code */
}