Intel® Fortran Compiler 17.0 Developer Guide and Reference
Enables or disables OpenMP* offloading compilation for the TARGET directives. This option only applies to Intel® MIC Architecture. Option -qopenmp-offload is the replacement option for -openmp-offload, which is deprecated.
Only available on Intel® 64 architecture targeting Intel® MIC Architecture
Linux: | -qopenmp-offload[=device] -qno-openmp-offload |
macOS: | None |
Windows: | /Qopenmp-offload[:device] /Qopenmp-offload- |
device |
Specifies the default device for target pragmas. Possible values are:
|
None
-qno-openmp-offload or /Qopenmp-offload- |
OpenMP* offloading compilation is disabled. However, if option [q or Q]openmp is specified, the default is ON and OpenMP offloading compilation is enabled. |
This option enables or disables OpenMP* offloading compilation for the TARGET directives. When enabling offloading, it lets you specify what the default target device should be for the TARGET directives. The default is mic.
You can also use this option if you want to enable or disable the offloading feature with no impact on other OpenMP* features. In this case, no OpenMP runtime library is needed to link and the compiler does not need to generate OpenMP runtime initialization code.
If you specify this option with the [q or Q]openmp option, it can impact other OpenMP* features.
None
Consider the following:
-qno-openmp -qopenmp-offload ! Linux /Qopenmp- /Qopenmp-offload ! Windows
The above is equivalent to specifying only [q or Q]openmp-offload. In this case, only the offload library is linked, not the OpenMP* library, and only the !$OMP directives for TARGET are processed but no other !$OMP directives.
Consider the following:
-qopenmp -qopenmp-offload ! Linux /Qopenmp /Qopenmp-offload ! Windows
In this case, the offload library is linked, the OpenMP library is linked, and OpenMP runtime initialization code is generated.