[Bug tree-optimization/80738] dead first stmt in a=0;a=b;b=0 whatever the aliasing

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80738

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2019-06-18
     Ever confirmed|0                           |1

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Btw, what's a C++ testcase for this?  With a simple

#include <utility>

int f(int i, int j)
{
  std::swap(i, j);
  return i;
}

I don't see this pattern.  If it only happens with more complex types
the fact that we may end up with aggregate copies or multiple loads/stores
will break the simplistic pattern detection.