Re: Proper Tail Recursive C++

William D Clinger <will@ccs.neu.edu> writes:


      int f (int n) {
              int A[1000];
              A[0] = n;
              A[999] = n;
              return g (A); // tail-recursive call to g
      }
[...]
      It is extremely unlikely that any C++ compiler is properly tail
      recursive in this sense.


Well, since no one else tried it: g++ compiles this, but the result
crashes. I gather that this is because it assumes that A is dead
before the call to g(). This permits it the luxury of performing the
tail recursion optimization, at the expense of producing code that
doesn't work :-)


Without either global analysis or garbage collection, I don't see any
way for the example to be made to work in C++, since the lifetime of
the automatic variable has to exceed the life of a tail recursion
optimized procedure. One could call this a bug, unless it were
documented as a feature.


nathan
[Tail recursion optimization does indeed seem to be of interest mostly to
users of languages like Scheme that promise garbage collection. -John]


--

Return to the comp.compilers page.
Search the comp.compilers archives again.