When compiling with ifort before version 17.0.1, the default behaviour is for the compiler not to use the Fortran rules for automatic allocation on intrinsic assignment. You will need to use an option like -assume realloc-lhs.
In years past, ignoring this use of the first column could cause bad things on a line printer, like page ejects — but when was the last time that you saw a line printer? Otherwise it was output device, compiler and OS dependent. The advice of “Fortran 95/2003 for Scientists and Engineers” was excellent for … Read more
To add to Vladimir F’s answer I’ll mention that gfortran 5.0 (but not earlier) supports the IEEE intrinsic modules. Instead of real x x=0 x=0/x one can use use, intrinsic :: iso_fortran_env use, intrinsic :: ieee_arithmetic integer(int32) i real(real32) x x = ieee_value(x, ieee_quiet_nan) i = transfer(x,i) This gives you a little flexibility over which … Read more
You can tell gfortran where your module files (.mod files) are located with the -I compiler flag. In addition, you can tell the compiler where to put compiled modules with the -J compiler flag. See the section “Options for directory search” in the gfortran man page. I use these to place both my object (.o … Read more
You are correct that m31tensorprod being an internal function means that you do not have to declare it in the main program. In the jargon: it has an explicit interface. However, that is not the problem with your code. What is going wrong is with the function definition itself. [Admittedly the compiler message isn’t too … Read more
Is it currently possible to override the structure constructor in Fortran? No. Anyway even using your approach is completely not about constructor overriding. The main reason is that structure constructor # OOP constructor. There is some similarity but this is just another idea. You can not use your non-intrinsic function in initialization expression. You can … Read more
As already answered, you need to specify widths. Something like ES14.5 might work well for the floating point format. There is a short cut for the integer format: I0 will cause the compiler to use the number of digits needed.
With gcc, you need to compile and link with -fopenmp to enable OpenMP. Other compilers have different options; with intel it’s -openmp, with pgi it’s -mp, etc.
Use (gdb) print *((real *)A+m)@n where A is the array, m is the shift in the position (no need to write +m in the case m=0) and n is the number of elements that you want to print. If you’re working with double precision reals then replace real by real_8. Also, for integers replace real … Read more
You need to modify the ~/.R/Makevars file. For a greater overview of this see: https://cran.r-project.org/doc/manuals/r-release/R-admin.html#OS-X-packages Alternatively, this has been answered before in a bit more depth by @kevin-ushey in Rcpp warning: “directory not found for option ‘-L/usr/local/Cellar/gfortran/4.8.2/gfortran’”. What is happening is your code is not being run under gcc instead it is being forwarded to … Read more