------- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-24 22:02 ------- Hmm, with my example in comment #5, this is a regression from 2.95.3 and not from 3.4.0, I should have really tested it before saying what I thought cause the regression.
The patch which I am thinking caused this regression now is: 2000-04-08 Mark Mitchell <[EMAIL PROTECTED]> * cp-tree.h (cp_tree_index): Add CPTI_COMPLETE_CTOR_IDENTIFIER. (complete_ctor_identifier): New macro. (special_function_kind): Add sfk_copy_constructor and sfk_assignment_operator. It is not setting the locus correctly on the body of the functions. -- What |Removed |Added ---------------------------------------------------------------------------- Known to fail| |3.3.3 3.4.0 4.0.0 Known to work| |2.95.3 Summary|[4.0 Regression] Wrong line |[3.3/3.4/4.0 Regression] |number in diagnostic with |Wrong line number in |gcc 4.0 |diagnostic with gcc 4.0 Target Milestone|4.0.0 |3.4.4 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20164