[Bug c++/14639] [3.5 Regression] [non-unit-at-a-time] Incorrect emission of unused compiler-generated destructor

bowdidge at apple dot com gcc-bugzilla@gcc.gnu.org
Fri Jun 4 02:12:00 GMT 2004


------- Additional Comments From bowdidge at apple dot com  2004-06-04 02:12 -------
The test case passes fine, but the first compilation in the Qt installation gets stuck in an infinite loop in 
finish_file.  The first part of the patch (in passes.c) is in rest_of_decl_compilation, so I'd guess that's the 
cause.   I'll try backing out that change, and see what happens.

Here's part of the sample:

    248 Thread_110b
      248 start
        248 _start
          248 toplev_main
            248 do_compile
              248 compile_file
                248 c_common_parse_file
                  248 finish_file
                    248 walk_namespaces_r
                      239 wrapup_global_declarations
                        150 wrapup_global_declarations
                        72 rest_of_decl_compilation
                          39 cgraph_varpool_finalize_decl
                            25 cgraph_varpool_node
                              13 htab_find_slot_with_hash
                                11 htab_find_slot_with_hash
                                2 eq_varpool_node
                                  2 eq_varpool_node
                              7 cgraph_varpool_node
                              3 htab_find_slot
                                3 htab_find_slot


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14639



More information about the Gcc-bugs mailing list