[Bug c++/12491] [3.3/3.4 Regression] [eh] Destructor fails to compile when optimizations (inlining) are enabled

[bangerth at dealii dot org]

------- Additional Comments From bangerth at dealii dot org  2003-12-06 21:12 -------
If I remember correctly, then in-charge and not-in-charge refers to whether this particular 
destructor needs to call the destructors of virtual base classes. They should be the 
same if a class has no virtual base classes. 
 
The fact that gcc can't merge them if they are identical had, at least in the past, to do with 
the fact that the two copies have different mangled names, so we would have multiple 
entry points for the same function if merged (to be correct, the entry point for the two 
functions would be the same in this case). 
 
W. 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12491