OpenMP-Environment

 view release on metacpan or  search on metacpan

benchmarks/NPB3.4.1/NPB3.4-OMP/DC/README  view on Meta::CPAN

   'make.def' in the config directory and modify the file if necessary
   (an example of make.def provided in DC directory). 
   Then
      make dc CLASS=S

   If a compiler complains about type 'int64' already defined, add
   "-DHAS_INT64" to the CFLAGS list in make.def.

2. OpenMP environment needs to be set before program can be executed.
   First set the number of the threads:
   setenv OMP_NUM_THREADS 4
   Then to fix OpenMP implemantations on some machines:
   limit stacksize unlimit
   If running on Altix 
   setenv KMP_MONITOR_STACKSIZE 50m

3. Run
   A text file ADC.par is used to set DC parameters when the class 
   is undefined (U). 
   The file has 5 lines. The lines with 'key' words attrNum, measuresNum, 
   and tuplesNum define the number of dimensions, measures,
   and input tuples respectively. There a special parameter INVERSE_ENDIAN
   allows us to create data in non-native endian format (INVERSE_ENDIAN=1). 
   The last parameter(fileName) specifies a DC file set name, including
   (optionally) a full path to a directory which will contain all

benchmarks/NPB3.4.1/NPB3.4-OMP/README.install  view on Meta::CPAN

     1. Similar to randdp but written to be easier to vectorize. 
   

2. Execution

   The executable is named <benchmark-name>.<class>.x and is placed
   in the bin subdirectory (or in the directory BINDIR specified in
   make.def, if you've defined it).  Folllowing is an example of running 
   a benchmark in csh:

      setenv OMP_NUM_THREADS 4
      bin/bt.A.x > BT.A_out.4

   It runs BT Class A problem on 4 threads and the output is stored
   in BT.A_out.4.

   Each benchmark includes a set of additional timers for profiling purpose
   (reporting timing for selected code blocks).  By default, these timers
   are disabled.  To enable the timers, set the environment variable
   NPB_TIMER_FLAG to one of:

benchmarks/NPB3.4.1/NPB3.4-OMP/README.install  view on Meta::CPAN

   OpenMP 3.0.  Thus, the 3.4 version requires a compiler that supports
   OpenMP 3.0.

   NPB-OMP assumes 'deterministic' static scheduling at run-time to 
   ensure the correctness of the results.  Verification in some
   benchmarks might fail if this condition is not met. 

   For larger problem sizes, the default stack size for slave threads
   may need to be increased on certain platforms.  For OpenMP 3.0-compliant
   compilers, use the runtime environment variable:
      setenv OMP_STACKSIZE 50m  (for 50MB)

   In order to build the class E version of CG, the integer type
   needs to be promoted to 64-bit, which is usually done through 
   compilation flag (such as "-i8" for FFLAGS in config/make.def).

4. Notes on the implementation

   - Based on NPB3.0-SER, except that FT was kept closer to
     the original version in NPB2.3-serial.

benchmarks/NPB3.4.1/NPB3.4-OMP/test_scripts/runit  view on Meta::CPAN

module list >>& $logfile
echo "" >> $logfile

set cnt=0
set cntf=0
set cntp=0

set aps=(bt sp lu lu ua ua)
set spv=(blk blk hp doac au rd)
set c="A"
setenv NPB_TIMER_FLAG 1

foreach nt (4)
foreach cf (gcc itc_p pgi)

set bindir=bin_$cf
set outdir=out_$cf
if ( ! -d $outdir) mkdir -p $outdir

foreach ap (bt cg ep ft is lu mg sp ua)
   set pgm=${ap}.${c}.x