Visible to Intel only — GUID: GUID-41AEA1BE-F452-4A0A-8317-7B74B2D9464B
Visible to Intel only — GUID: GUID-41AEA1BE-F452-4A0A-8317-7B74B2D9464B
fopenmp-targets, Qopenmp-targets
Enables offloading to a specified GPU target if OpenMP* features have been enabled. This feature is only available for ifx.
Syntax
Linux: |
-fopenmp-targets=triple |
Windows: |
/Qopenmp-targets:triple |
Arguments
triple |
Is a target triple device name. The following triplets are supported.
For example, when you specify spir64, the compiler generates an x86 + SPIR64 (64-bit Standard Portable Intermediate Representation) fat binary for Intel® GPU devices. |
Default
OFF |
If this option is not specified, no fat binaries are created. |
Description
This option enables offloading to a specified GPU target if OpenMP* features have been enabled.
To use this option, you must enable recognition of OpenMP* features by specifying one of the following options:
Linux
-qopenmp
-fiopenmp (ifx only)
-fopenmp
Windows
/Qopenmp
/Qiopenmp (ifx only)
The following shows an example:
ifx -qopenmp -fopenmp-targets=spir64 matmul_offload.cpp -o matmul
When you specify -fopenmp-targets or /Qopenmp-targets, C++ exception handling is disabled for target compilations.
Linux
For host compilations, if you want to disable C++ exception handling, you must specify option -fno-exceptions.
Files compiled with option -fopenmp-targets or /Qopenmp-targets should also be linked with the same option, or the link step may fail.
When OpenMP* offloading is enabled, this option only applies to device-specific compilation.
IDE Equivalent
Visual Studio: Code Generation > Enable OpenMP Offloading
Alternate Options
None