Absolutely. Quite a lot of quantum chemistry here codes are Fortran, and most would use Intel Fortran for performance. While some (VASP) might depend depend on -heap-arrays Intel switch being used with a small value, the default setting for Intel Fortran is -no-heap-arrays "temporary arrays are allocated on the stack (DEFAULT)". Which leads potentially to the code using large stack, way over 10Mb.
-- Grigory Shamov Westgrid/ComputeCanada Site Lead University of Manitoba E2-588 EITC Building, (204) 474-9625 On 15-09-30 5:19 AM, "users on behalf of Dave Love" <users-boun...@open-mpi.org on behalf of d.l...@liverpool.ac.uk> wrote: >Nathan Hjelm <hje...@lanl.gov> writes: > >> I would like to add that you may want to play with the value and see >> what works for your applications. Most applications should be using >> malloc or similar functions to allocate large memory regions in the heap >> and not on the stack. > >It's long been a Fortran optimization to stack-allocate arrays (from the >days of all-the-world's-a-VAX when loads of code broke that assumed SAVE >attributes). It's an FAQ here for the intel compiler and well-optimized >gfortran, and there are other good reasons for large stacks compared >with the 10MB-ish default. >_______________________________________________ >users mailing list >us...@open-mpi.org >Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >Link to this post: >http://www.open-mpi.org/community/lists/users/2015/09/27724.php
