PetscRealConstant#
a compile time macro that ensures a given constant real number is properly represented in the configured precision of PetscReal be it half, single, double or 128-bit representation
Synopsis#
#include <petscmath.h>
PetscReal PetscRealConstant(real_number)
Not Collective
Input Parameter#
v1 - the real number, for example 1.5
Note#
For example, if PETSc is configured with --with-precision=__float128 and one writes
PetscReal d = 1.5;
the result is 1.5 in double precision extended to 128 bit representation, meaning it is very far from the correct value. Hence, one should write
PetscReal d = PetscRealConstant(1.5);
See Also#
Level#
beginner
Location#
Index of all Sys routines
Table of Contents for all manual pages
Index of all manual pages