PETSC will compile a special version of Blas/Lapack for long with-precision=longdouble -download-c-blas-lapack=1 To support long double, configure PETSC with additional To achieve better convergence rate and more accurate result. GSS can be compiled with long double float number (default is Petsc must be configured with -with-clanguage=cxx to supportĪnd -with-mpi=0 to disable parallel mechanism. Your $PETSC_DIR/src/snes/impls/ls and replace the existing file. I had reported it to PETSC development term, Petsc-2.3.3 which contains basic nonlinear and linear solvers. You should install two important libraries first. Please use the latest version of flex and bison. Inefficient way but icc seems using an hardware float point accelerator. Mainly because the math library used by gcc compute pow(x,y) in an (such as Intel c/c++ on PC) works much faster (50% faster than gcc). Although gcc 3.3/3.4/4.1/4.2 works, but vendor compiler Please make sure your c/c++ compiler works, flex-2.5.4 and bison-2.0 are also It is reported that GSS has been successfully Can be installed on Linux/Unix system and Cygwin under Windows
0 Comments
Leave a Reply. |