id	summary	reporter	owner	description	type	status	milestone	component	version	severity	resolution	keywords	cc
5143	segfaults in ordered_index	christoph.kluge@…	Joaquín M López Muñoz	"Hi,

we are suffering segfaults in
ordered_index_node_impl::increment line 254

static void increment(pointer& x)
{
  if(x->right()!=pointer(0)){
    x=x->right();
    while(x->left()!=pointer(0))x=x->left(); // XXX
  }
  else{
    pointer y=x->parent();
    while(x==y->right()){
      x=y;
      y=y->parent();
    }
    if(x->right()!=y)x=y;
  }
}

What, if x->left() becomes 0? Well, it does, over here! Shouldn't this
read

    while(x!=pointer(0)&&x->left()!=pointer(0))x=x->left();

? What is this line of code supposed to do anyway, other than
uselessly consuming CPU time? Why not write

    x = 0;

instead, if this was meant? ordered_index_node_impl_base::left and
ordered_index_node_impl_base::right do not have any side-effects.

By the way, for the classes ordered_index_node_std_base and
ordered_index_node_compressed_base, I don't see any of the POD data
members getting initialized. A user-provided constructor is not
defined. From the C++ Standard, IIRC, POD data members are not
initialized by the implicitly-generated constructor and therefore have
indeterminate values.

This also applies to recent versions of boost, the trunk included. The
code doesn't appear to have changed here.

To me this looks like this is working only by chance. Has anyone ever
bothered to review this code?

Cheers,
 Christoph

"	Bugs	closed	To Be Determined	multi_index	Boost 1.35.0	Problem	invalid