Previous: glpk errors, Up: glpk


D.4.4 Additional glpk notes

A sparse matrix representation of m is used because in practice most linear programming problems have very sparse systems of equations.

Only the constraint of non-negativity is admitted. Other constraints such as upper bounds must be effected through a change of variables if required.

The glpk library has a small memory leak, which avram corrects by methods described in Memory leaks.