FLANG
|
#include <flang/Support/LangOptions.h>
Public Attributes | |
std::string | OMPHostIRFile |
signed integer overflow handling | |
std::vector< llvm::Triple > | OMPTargetTriples |
List of triples passed in using -fopenmp-targets. |
Additional Inherited Members | |
Public Types inherited from Fortran::common::LangOptionsBase | |
enum | SignedOverflowBehaviorTy { SOB_Undefined , SOB_Defined } |
enum | FPModeKind { FPM_Off , FPM_Fast } |
Tracks various options which control the dialect of Fortran that is accepted. Based on clang::LangOptions
std::string Fortran::common::LangOptions::OMPHostIRFile |
signed integer overflow handling
< FP Contract Mode (off/fast) Indicate a build without the standard GPU libraries. Permit floating point optimization without regard to infinities Permit floating point optimization without regard to NaN Allow math functions to be replaced with an approximately equivalent calculation Allow optimizations that ignore the sign of floating point zeros Allow reassociation transformations for floating-point instructions Allow division operations to be reassociated Specify version of OpenMP standard Generate code only for OpenMP target device Generate OpenMP target code only for GPUs Generate OpenMP target code only for GPUs Enable debugging in the OpenMP offloading device RTL Assume work-shared loops do not have more iterations than participating threads. Assume distributed loops do not have more iterations than participating teams. Assume that no thread in a parallel region will modify an ICV. Assume that no thread in a parallel region will encounter a parallel region Use SIMD only OpenMP support. < Minimum vscale range value Maximum vscale range value Name of the IR file that contains the result of the OpenMP target host code generation.