[Bug lto/46911] New: [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[Bug lto/46911] New: [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[zsojka at seznam dot cz]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

           Summary: [4.6 Regression] ICE: SIGSEGV in
                    add_name_and_src_coords_attributes (dwarf2out.c:17792)
                    with -flto -g
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: lto
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: zsojka@seznam.cz
              Host: x86_64-pc-linux-gnu
            Target: x86_64-pc-linux-gnu


----- testcase.f -----
      common/main1/ eps(2)
      call dalie6s(iqmod6,1,wx,cor6d)
      end
----------------------
reduced from gfortran.dg/lto/20091015-1_2.f

Compiler output:
$ gfortran -O -finline-small-functions -flto -g testcase.f -r -nostdlib
==5279== Invalid read of size 2
==5279==    at 0x5C75BC: add_name_and_src_coords_attributes (dwarf2out.c:17792)
==5279==    by 0x5D8104: gen_variable_die (dwarf2out.c:19400)
==5279==    by 0x5DCF14: gen_decl_die (dwarf2out.c:20981)
==5279==    by 0x5DA34A: decls_for_scope (dwarf2out.c:20555)
==5279==    by 0x5DF38B: gen_block_die (dwarf2out.c:19669)
==5279==    by 0x5DA37D: decls_for_scope (dwarf2out.c:20567)
==5279==    by 0x5DF582: gen_block_die (dwarf2out.c:19701)
==5279==    by 0x5DA37D: decls_for_scope (dwarf2out.c:20567)
==5279==    by 0x5DA8FF: gen_subprogram_die (dwarf2out.c:19277)
==5279==    by 0x5DCDA2: gen_decl_die (dwarf2out.c:20914)
==5279==    by 0x5E00E8: dwarf2out_function_decl (dwarf2out.c:21296)
==5279==    by 0x638FCC: rest_of_handle_final (final.c:4286)
==5279==    by 0x771ADE: execute_one_pass (passes.c:1553)
==5279==    by 0x771DA4: execute_pass_list (passes.c:1608)
==5279==    by 0x771DB6: execute_pass_list (passes.c:1609)
==5279==    by 0x771DB6: execute_pass_list (passes.c:1609)
==5279==    by 0x8B2605: tree_rest_of_compilation (tree-optimize.c:422)
==5279==    by 0xA77EB1: cgraph_expand_function (cgraphunit.c:1508)
==5279==    by 0xA7A489: cgraph_optimize (cgraphunit.c:1567)
==5279==    by 0x4FF78E: lto_main (lto.c:2453)
==5279==    by 0x85BDE7: toplev_main (toplev.c:579)
==5279==    by 0x6369BBC: (below main) (in /lib64/libc-2.11.2.so)
==5279==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==5279== 
In file included from :0:0:
testcase.f: In function 'main':
testcase.f:3:0: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
lto-wrapper:
/mnt/svn/gcc-trunk/binary-167643-lto-fortran-checking-yes-rtl-df/bin/gfortran
returned 1 exit status
collect2: lto-wrapper returned 1 exit status

Tested revisions:
r167643 - crash
r165699 - crash
r161659 - OK
4.5 r166509 - OK

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.6.0

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P4
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011.01.18 17:38:15
                 CC|                            |jakub at gcc dot gnu.org
     Ever Confirmed|0                           |1

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-01-18 17:38:15 UTC ---
When dwarf2out_abstract_function is called on MAIN__, it doesn't have
DECL_INITIAL set when in lto1 (compared to f951), the BLOCK tree is simply not
present, which both causes incomplete debug information and for Fortran COMMON
blocks ICEs, as it relies on those to be actually present.  P4 as it only
affects Fortran, but if -flto -g is ever meant to be usable, the BLOCK trees
need to be streamed out and preserved.  It seems output_function calls:
  /* Output DECL_INITIAL for the function, which contains the tree of
     lexical scopes.  */
  lto_output_tree (ob, DECL_INITIAL (function), true);
but for inlines output_function isn't called.

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-19 16:53:11 UTC ---
(In reply to comment #1)
> When dwarf2out_abstract_function is called on MAIN__, it doesn't have
> DECL_INITIAL set when in lto1 (compared to f951), the BLOCK tree is simply not
> present, which both causes incomplete debug information and for Fortran COMMON
> blocks ICEs, as it relies on those to be actually present.  P4 as it only
> affects Fortran, but if -flto -g is ever meant to be usable, the BLOCK trees
> need to be streamed out and preserved.  It seems output_function calls:
>   /* Output DECL_INITIAL for the function, which contains the tree of
>      lexical scopes.  */
>   lto_output_tree (ob, DECL_INITIAL (function), true);
> but for inlines output_function isn't called.

That's true.  Inlines would be reachable via the block trees abstract
origins, right?  We do stream that, so the only thing missing is
associating it with the function-decl ... (maybe we can do that
when reading the BLOCK and its context is a FUNCTION_DECL?)

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-01-19 16:55:55 UTC ---
Yeah, all the inlines should be BLOCK_ABSTRACT_ORIGIN of some BLOCK that is
being streamed, so in theory all of them should be already streamed, just
DECL_INITIAL is not filled in.

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

--- Comment #4 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-19 17:10:31 UTC ---
We already do

  /* We re-compute BLOCK_SUBBLOCKS of our parent here instead
     of streaming it.  For non-BLOCK BLOCK_SUPERCONTEXTs we still
     stream the child relationship explicitly.  */
  if (BLOCK_SUPERCONTEXT (expr)
      && TREE_CODE (BLOCK_SUPERCONTEXT (expr)) == BLOCK)
    {
      BLOCK_CHAIN (expr) = BLOCK_SUBBLOCKS (BLOCK_SUPERCONTEXT (expr));
      BLOCK_SUBBLOCKS (BLOCK_SUPERCONTEXT (expr)) = expr;
    }

so adding a

  if (BLOCK_SUPERCONTEXT (expr)
      && TREE_CODE (BLOCK_SUPERCONTEXT (expr)) == FUNCTION_DECL)
    DECL_INITIAL (BLOCK_SUPERCONTEXT (expr)) = expr;

shouldn't be too bad - but it does not fixe the testcase ... (but
DECL_INITIAL is now filled in correctly - I bet without VAR_DECLs
though).

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Blocks|                            |47819

--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-02-23 12:06:34 UTC ---
Same happens for 416.gamess.

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

--- Comment #6 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-02-23 13:34:25 UTC ---
There are several issues.  First as Jakub noted we have a NULL DECL_INITAL
for the inlined __MAIN.  If we fix that during LTO streaming (avoiding direct
input/output of DECL_INITIAL of all FUNCTION_DECLs) we run into the issue
that cgraph sets DECL_INITIAL of __MAIN to error_mark_node after inlining
it (assuming that debug_hooks->outlining_inline_function did everything that
was necessary already).  We can try to avoid this by setting
abstract_and_needed in the cgraph node.

Together this solves this ICE.  It doesn't solve PR47799 though.

This is because how we split trees between the global and the function-local
sections during streaming.  We end up with multiple copies of the abstract
BLOCK tree as well as bogus cross-function references for all abstract
origins (apart from that to the function-decl itself).

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

--- Comment #7 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-02-23 14:35:27 UTC ---
Created attachment 23443
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23443
prototype patch

Patch also trying to fix PR47799 (but we'll still have multiple copies of
the BLOCK trees).  For better clarity the BLOCK_ABSTRACT_ORIGIN BLOCK
trees should probably be streamed from input/output_function by walking
the functions DECL_INITIAL.  That would make it clear that the BLOCKs
are duplicated in each caller function section (and that we mess up
DECL_INITIAL of the abstract origin function-decls for all but the
last function).

We probably need to output the abstract function and its BLOCK tree to
the global decl&type LTO section, possibly triggered by
debug_hooks->outlining_inline_function.

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

--- Comment #8 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-03-01 09:45:09 UTC ---
Author: rguenth
Date: Tue Mar  1 09:45:05 2011
New Revision: 170588

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=170588
Log:
2011-03-01  Richard Guenther  <rguenther@suse.de>

    PR lto/46911
    * lto-streamer-in.c (lto_input_ts_decl_common_tree_pointers):
    Do not stream DECL_ABSTRACT_ORIGIN.
    (lto_input_ts_block_tree_pointers): Nor BLOCK_SOURCE_LOCATION,
    BLOCK_NONLOCALIZED_VARS or BLOCK_ABSTRACT_ORIGIN.
    * lto-streamer-out.c (lto_output_ts_decl_common_tree_pointers):
    Do not stream DECL_ABSTRACT_ORIGIN.
    (lto_output_ts_block_tree_pointers): Nor BLOCK_SOURCE_LOCATION,
    BLOCK_NONLOCALIZED_VARS or BLOCK_ABSTRACT_ORIGIN.

    * gfortran.dg/lto/pr46911_0.f: New testcase.

Added:
    trunk/gcc/testsuite/gfortran.dg/lto/pr46911_0.f
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/lto-streamer-in.c
    trunk/gcc/lto-streamer-out.c
    trunk/gcc/testsuite/ChangeLog

[Bug lto/46911] [4.6 Regression] ICE: SIGSEGV in add_name_and_src_coords_attributes (dwarf2out.c:17792) with -flto -g

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46911

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

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

--- Comment #9 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-03-01 09:46:36 UTC ---
Fixed.