[Bug c++/21006] New: [4.1 Regression] g++.dg/other/static11.C fails

[Bug c++/21006] New: [4.1 Regression] g++.dg/other/static11.C fails

[jsm28 at gcc dot gnu dot org]

FAIL: g++.dg/other/static11.C (test for excess errors)

has appeared on mainline, apparently on all targets, between 20050412 and 20050413.

On i686-pc-linux-gnu:

/home/gcc/nightlies/gcc-mainline-2005-04-13/gcc/testsuite/g++.dg/other/static11.C:
In constructor 'C<T>::C() [with T = int]':
/home/gcc/nightlies/gcc-mainline-2005-04-13/gcc/testsuite/g++.dg/other/static11.C:21:
internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

-- 
           Summary: [4.1 Regression] g++.dg/other/static11.C fails
           Product: gcc
           Version: 4.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: jsm28 at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-04-13 21:37 -------
confirmed.  This is more likely a middle-end problem, the options are "-da".

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
          Component|c++                         |middle-end
     Ever Confirmed|                            |1
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2005-04-13 21:37:37
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-04-13 21:38 -------
http://gcc.gnu.org/ml/gcc-regression/2005-04/msg00051.html
Has the changelog of who might have caused this.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.1.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[hp at gcc dot gnu dot org]

------- Additional Comments From hp at gcc dot gnu dot org  2005-04-14 02:01 -------
Fails for cris-elf and mmix-knuth-mmixware as well.  Timespan that narrows down
the suspects a little: worked on LAST_UPDATED "Tue Apr 12 20:48:11 UTC 2005",
started failing "Wed Apr 13 05:00:11 UTC 2005".

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hp at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-04-15 15:55 -------
Also I can confirm that this is a middle-end problem.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pinskia at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[jsm28 at gcc dot gnu dot org]

------- Additional Comments From jsm28 at gcc dot gnu dot org  2005-04-18 11:20 -------
Appeared on x86_64-linux between 2005-04-12 21:33 UTC and 2005-04-12 21:37 UTC.
 I.e., caused by

2005-04-12  Steven Bosscher  <stevenb@suse.de>
            Stuart Hastings <stuart@apple.com>
            Jan Hubicka  <jh@suse.cz>

        * Makefile.in: Add function.h to BASIC_BLOCK_H.  Remove all
        references to gt-tree-cfg.h.
        * basic-block.h (struct basic_block_def): Don't skip rbi
        for garbage collection.
        (struct reorder_block_def): Make GTY-able.
        (struct control_flow_graph): New structure.
        (n_edges, n_basic_blocks, last_basic_block, basic_block_info,
        BASIC_BLOCK, EXIT_BLOCK_PTR, ENTRY_BLOCK_PTR): No longer vars,
        but instead defines to the control_flow_graph for cfun.
        (label_to_block_map): New define, points to the label map of
        the control_flow_graph for cfun.
        (n_edges_for_function, n_basic_blocks_for_function,
        last_basic_block_for_function, basic_block_info_for_function,
        EXIT_BLOCK_PTR_FOR_FUNCTION, ENTRY_BLOCK_PTR_FOR_FUNCTION,
        basic_block_info_for_function, label_to_block_map_for_function):
        Counterparts for the above, taking a struct function as an extra
        argument.
        (alloc_rbi_pool, free_rbi_pool): Remove prototypes.
        * cfg.c: (n_edges, n_basic_blocks, last_basic_block,
        basic_block_info, ENTRY_BLOCK_PTR, EXIT_BLOCK_PTR): Remove.
        (alloc_rbi_pool, free_rbi_pool): Remove.
        (initialize_bb_rbi): Use ggc_alloc_cleared instead of pool_alloc.
        * cfglayout.c: (cfg_layout_initialize): Don't allocate the rbi pool
        here...
        (cfg_layout_finalize) ... and don't free it here.
        * cfgrtl.c (cfg_layout_delete_block): Zero out rbi so it gets
        garbage collected.
        * flow.c (free_basic_block_vars): Set label_to_block_map and
        n_edges to zero too.
        * function.h (struct function): Add cfg field.
        * function.c (allocate_struct_function): Allocate the cfg.
        * tree-cfg.c (label_to_block_map): Remove.
        (build_tree_cfg): Don't allocate the rbi pool here...
        (delete_tree_cfg_annotations): ...and don't free it here.
        Also don't nullify label_to_block_map for cfun.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |steven at gcc dot gnu dot
                   |                            |org, stuart at apple dot
                   |                            |com, hubicka at gcc dot gnu
                   |                            |dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[jsm28 at gcc dot gnu dot org]

------- Additional Comments From jsm28 at gcc dot gnu dot org  2005-05-04 19:40 -------
Appears to have started passing again (on hppa2.0w-hpux, hppa64-hpux,
i686-linux,  ia64-hpux at least) between 20050503 and 20050504.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006

[Bug middle-end/21006] [4.1 Regression] g++.dg/other/static11.C fails

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-05-04 20:07 -------
Fixed by:
2005-05-03  Richard Henderson  <rth@redhat.com>
        * cfg.c (dump_flow_info): Use max_reg_num, not max_regno.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21006