about upc-benchmark

• Previous message: Paul H. Hargrove: "Re: upc-benchmark"

Hi,

  The platform is DSMP, and the conduit is mpi, the make.def under the
dirction of config is:

#---------------------------------------------------------------------------

# This is the UPC compiler used for UPC programs

#---------------------------------------------------------------------------

CC = upcc

# This links C programs; usually the same as ${CC}

CLINK   = upcc

 

#---------------------------------------------------------------------------

# These macros are passed to the linker

#---------------------------------------------------------------------------

C_LIB  = -lm -T ${NP} -network=mpi

 

#---------------------------------------------------------------------------

# These macros are passed to the compiler

#---------------------------------------------------------------------------

C_INC = -I../common

 

#---------------------------------------------------------------------------

# Global *compile time* flags for C programs

#---------------------------------------------------------------------------

CFLAGS  = -T ${NP} -O -network=mpi -Wc,-g

 

#---------------------------------------------------------------------------

# Global *link time* flags. Flags for increasing maximum executable

# size usually go here.

#---------------------------------------------------------------------------

CLINKFLAGS =

 

#---------------------------------------------------------------------------

# Utilities C:

#

# This is the C compiler used to compile C utilities.  Flags required by

# this compiler go here also; typically there are few flags required; hence

# there are no separate macros provided for such flags.

#---------------------------------------------------------------------------

UCC     = upcc -T ${NP} -O -network=mpi -Wc,-g

 

#---------------------------------------------------------------------------

# Destination of executables, relative to subdirs of the main directory. .

#---------------------------------------------------------------------------

BINDIR  = ../bin

 

 

#---------------------------------------------------------------------------

# The variable RAND controls which random number generator

# is used. It is described in detail in Doc/README.install.

# Use "randi8" unless there is a reason to use another one.

# Other allowed values are "randi8_safe", "randdp" and "randdpvec"

#---------------------------------------------------------------------------

RAND   = randi8

# The following is highly reliable but may be slow:

#RAND   = randdp

 

 

#---------------------------------------------------------------------------

# The variable WTIME is the name of the wtime source code module in the

# NPB2.x/common directory.

# For most machines,       use wtime.c

# For SGI power challenge: use wtime_sgi64.c

#---------------------------------------------------------------------------

WTIME  = wtime.c

 

 

#---------------------------------------------------------------------------

# Enable if either Cray or IBM:

# (no such flag for most machines: see common/wtime.h)

# This is used by the C compiler to pass the machine name to common/wtime.h,

# where the C/Fortran binding interface format is determined

#---------------------------------------------------------------------------

# MACHINE       =       -DCRAY

# MACHINE       =       -DIBM

 

 

But some erros show below:

 

autopar@gnode10:~/lxj/NPB-UPC-110404/CG> make CLASS=S NP=4

gmake[1]: Entering directory `/bwdata/autopar/lxj/NPB-UPC-110404/sys'

o setparams setparams.c -lm

gmake[1]: o: Command not found

gmake[1]: [setparams] Error 127 (ignored)

gmake[1]: Leaving directory `/bwdata/autopar/lxj/NPB-UPC-110404/sys'

../sys/setparams cg S 4

UPC Runtime error: program was compiled with 4 static threads, but executed
with 1 threads.

NOTICE: Before reporting bugs, run with GASNET_BACKTRACE=1 in the
environment to generate a backtrace. 

make: *** [config] Error 255

 

 

what it mean?  

 

• Previous message: Paul H. Hargrove: "Re: upc-benchmark"