diff --git a/fpm.toml b/fpm.toml
index 66ea2c3..7cc1a76 100644
--- a/fpm.toml
+++ b/fpm.toml
@@ -12,13 +12,14 @@ auto-tests = true
 auto-examples = true
 module-naming = false
 link = ["blas", "lapack"]
+#link = ["openblas"]
 
 [install]
-library = false
+library = true
 
 [fortran]
 implicit-typing = false
-implicit-external = false
+implicit-external = true
 source-form = "free"
 
 [dependencies]
diff --git a/src/AbstractVector.f90 b/src/AbstractVector.f90
index 94c81b5..e06a2dd 100644
--- a/src/AbstractVector.f90
+++ b/src/AbstractVector.f90
@@ -13,41 +13,41 @@ module AbstractVector
    contains
      private
 
-     !> Basic operations.
+  !> Basic operations.
      procedure, pass(from), public :: copy
      generic  , public             :: assignment(=) => copy
 
      procedure(abstract_zero), deferred, public :: zero
 
-     !> Scalar-vector product.
+  !> Scalar-vector product.
      procedure(abstract_scal), deferred, public :: scal
 
-     !> Vector-vector operations.
+  !> Vector-vector operations.
      procedure(axpby_interface), deferred, pass(self), public :: axpby
      procedure, pass(self), public :: add
      procedure, pass(self), public :: sub
 
-     !> Reduction operations.
+  !> Reduction operations.
      procedure(abstract_dot), pass(self), deferred, public :: dot
      procedure, pass(self), public                         :: norm
   end type abstract_vector
 
-  ! --> Abstract interfaces for the type-bound procedures.
+  !> Abstract interfaces for the type-bound procedures.
   abstract interface
-     !> Basic operations.
+  !> Basic operations.
      subroutine abstract_zero(self)
        import abstract_vector
        class(abstract_vector), intent(inout) :: self
      end subroutine abstract_zero
 
-     !> Scalar-vector product.
+  !> Scalar-vector product.
      subroutine abstract_scal(self, alpha)
        import abstract_vector, wp
        class(abstract_vector), intent(inout) :: self
        real(kind=wp), intent(in)             :: alpha
      end subroutine abstract_scal
 
-     !> Vector-vector operations.
+  !> Vector-vector operations.
      subroutine axpby_interface(self, alpha, vec, beta)
        import abstract_vector, wp
        class(abstract_vector), intent(inout) :: self
@@ -55,7 +55,7 @@ subroutine axpby_interface(self, alpha, vec, beta)
        real(kind=wp)         , intent(in)    :: alpha, beta
      end subroutine axpby_interface
 
-     !> Reduction operations.
+  !> Reduction operations.
      real(kind=wp) function abstract_dot(self, vec) result(alpha)
        import abstract_vector, wp
        class(abstract_vector), intent(in) :: self, vec
@@ -67,6 +67,7 @@ end function abstract_dot
   subroutine copy(out, from)
     class(abstract_vector), intent(in)               :: from
     class(abstract_vector), allocatable, intent(out) :: out
+    if (allocated(out)) deallocate(out)
     allocate(out, source=from)
     return
   end subroutine copy
@@ -74,7 +75,7 @@ end subroutine copy
   subroutine add(self, y)
     class(abstract_vector), intent(inout) :: self
     class(abstract_vector), intent(in)    :: y
-    ! --> Vector addition.
+  !> Vector addition.
     call self%axpby(1.0_wp, y, 1.0_wp)
     return
   end subroutine add
@@ -82,7 +83,7 @@ end subroutine add
   subroutine sub(self, y)
     class(abstract_vector), intent(inout) :: self
     class(abstract_vector), intent(in)    :: y
-    ! --> Vector subtraction.
+  !> Vector subtraction.
     call self%axpby(1.0_wp, y, -1.0_wp)
     return
   end subroutine sub
@@ -93,29 +94,32 @@ real(kind=wp) function norm(self) result(alpha)
   end function norm
 
   subroutine get_vec(y, X, v)
-    !> Output vector.
+  !> Output vector.
     class(abstract_vector), allocatable, intent(out) :: y
-    !> Krylov basis.
+  !> Krylov basis.
     class(abstract_vector), intent(in) :: X(:)
-    !> Coordinates of the output vector y in the Krylov basis X.
+  !> Coordinates of the output vector y in the Krylov basis X.
     real(kind=wp), intent(in) :: v(:)
-    !> Temporary Krylov vector.
+  !> Temporary Krylov vector.
     class(abstract_vector), allocatable :: wrk
-    !> Miscellaneous.
+  !> Miscellaneous.
     integer :: i
-    ! --> Check sizes.
+  !> Check sizes.
     if (size(X) .ne. size(v)) then
        write(*, *) "INFO : Krylov basis X and low-dimension vector v have different sizes."
-       call exit()
+       return
     endif
-    ! --> Initialize output vector.
+  !> Initialize output vector.
     if (allocated(y) .eqv. .false.) allocate(y, source=X(1)) ; call y%zero()
-    ! --> Compute output vector.
+  !> Compute output vector.
+    if (.not. allocated(wrk)) allocate(wrk, source=X(1))
     do i = 1, size(X)
-       wrk = X(i) ; call wrk%scal(v(i))
+       wrk = X(i)
+       call wrk%scal(v(i))
        call y%add(wrk)
     enddo
+    if (allocated(wrk)) deallocate(wrk)
     return
   end subroutine get_vec
 
-end module AbstractVector
+end module AbstractVector
\ No newline at end of file