[Bug middle-end/67120] [6 Regression] wrong code for volatile pointers at -O1 and above on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67120

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Mine.  Happens in FRE.

Setting value number of a.1_4 to &c (changed)
Applying pattern match.pd:1961, generic-match.c:4683
Marking all edges out of BB 2 but (2 -> 3) as not executable
Marking all outgoing edges of unreachable BB 4 as not executable
...
Replaced a with &c in all uses of a.1_4 = a;
Applying pattern match.pd:1961, gimple-match.c:5236
gimple_simplified to if (1 != 0)