HPL_pdlamch man page

HPL_pdlamch — determines machine-specific arithmetic constants.

Synopsis

#include "hpl.h"
 double HPL_pdlamch( MPI_Comm COMM, const HPL_T_MACH CMACH );

Description

HPL_pdlamch determines  machine-specific  arithmetic  constants  such  as the relative machine precision (eps),  the safe minimum(sfmin) such that 1/sfmin does not overflow, the base of the machine (base), the precision (prec),  the  number  of  (base)  digits in the  mantissa  (t),  whether rounding occurs in addition (rnd = 1.0 and 0.0 otherwise),  the  minimum exponent before  (gradual)  underflow (emin),  the  underflow  threshold (rmin)- base**(emin-1), the largest exponent before overflow (emax), the overflow threshold (rmax)  - (base**emax)*(1-eps).

Arguments

COMM    (global/local input)    MPI_Comm

The MPI communicator identifying the process collection.

CMACH   (global input)          const HPL_T_MACH

Specifies the value to be returned by HPL_pdlamch            
  = HPL_MACH_EPS,   HPL_pdlamch := eps (default)            
  = HPL_MACH_SFMIN, HPL_pdlamch := sfmin                    
  = HPL_MACH_BASE,  HPL_pdlamch := base                    
  = HPL_MACH_PREC,  HPL_pdlamch := eps*base                
  = HPL_MACH_MLEN,  HPL_pdlamch := t                        
  = HPL_MACH_RND,   HPL_pdlamch := rnd                      
  = HPL_MACH_EMIN,  HPL_pdlamch := emin                    
  = HPL_MACH_RMIN,  HPL_pdlamch := rmin                    
  = HPL_MACH_EMAX,  HPL_pdlamch := emax                    
  = HPL_MACH_RMAX,  HPL_pdlamch := rmax                    
 where                                                        

  eps   = relative machine precision,                      
  sfmin = safe minimum,                                    
  base  = base of the machine,                              
  prec  = eps*base,                                        
  t     = number of digits in the mantissa,                
  rnd   = 1.0 if rounding occurs in addition,              
  emin  = minimum exponent before underflow,                
  rmin  = underflow threshold,                              
  emax  = largest exponent before overflow,                
  rmax  = overflow threshold.

Info

October 26, 2012 HPL 2.1 HPL Library Functions