[Bug c++/47074] New: tree check fail in create_access_replacement

[Bug c++/47074] New: tree check fail in create_access_replacement

[dcb314 at hotmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

           Summary: tree check fail in create_access_replacement
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: dcb314@hotmail.com


Created attachment 22855
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22855
gzipped C++ source code

I just tried to compile package freefem++-3.11 with the latest 4.6
snapshot 20101225 on a Linux x86_64 box. The compiler said

array_long.cpp:167:1: internal compiler error: tree check: expected tree that
contains 'decl minimal' structure, have 'pointer_type' in
create_access_replacement, at tree-sra.c:1679
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

Source code attached. Flag -O2 required.

[Bug c++/47074] [4.6 Regression] tree check fail in create_access_replacement

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jamborm at gcc dot gnu.org
   Target Milestone|---                         |4.6.0
            Summary|tree check fail in          |[4.6 Regression] tree check
                   |create_access_replacement   |fail in
                   |                            |create_access_replacement

[Bug c++/47074] [4.6 Regression] tree check fail in create_access_replacement

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> 2010-12-28 21:04:26 UTC ---
For me it ICEs in build_int_cst_wide, TYPE_CACHED_VALUES has been corrupted, so
it looks like a GC issue to me.

[Bug c++/47074] [4.6 Regression] tree check fail in create_access_replacement

[dcb314 at hotmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

--- Comment #2 from dcb <dcb314 at hotmail dot com> 2010-12-29 10:12:14 UTC ---
(In reply to comment #1)
> For me it ICEs in build_int_cst_wide, TYPE_CACHED_VALUES has been corrupted, so
> it looks like a GC issue to me.

gcc has been producing many valgrind / memcheck errors for the
last few weeks. This may or may not be related.

[Bug tree-optimization/47074] [4.6 Regression] tree check fail in create_access_replacement

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2010.12.29 13:51:25
         AssignedTo|unassigned at gcc dot       |jakub at gcc dot gnu.org
                   |gnu.org                     |
     Ever Confirmed|0                           |1

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> 2010-12-29 13:51:25 UTC ---
Created attachment 22865
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22865
gcc46-pr47074.patch

This fixes the ICE for me, can you see if it fixes it for you too?

[Bug tree-optimization/47074] [4.6 Regression] tree check fail in create_access_replacement

[dcb314 at hotmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

--- Comment #4 from dcb <dcb314 at hotmail dot com> 2010-12-29 17:43:29 UTC ---
(In reply to comment #3)
> Created attachment 22865 [details]
> gcc46-pr47074.patch
> 
> This fixes the ICE for me, can you see if it fixes it for you too?

I can confirm that the ice seems to have gone away for this 
test case only.

You'd have to check the entire gcc test suite to be sure, however.

[Bug tree-optimization/47074] [4.6 Regression] tree check fail in create_access_replacement

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-29 18:32:57 UTC ---
Looks obviously ok to me.

[Bug tree-optimization/47074] [4.6 Regression] tree check fail in create_access_replacement

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> 2010-12-29 21:49:40 UTC ---
Author: jakub
Date: Wed Dec 29 21:49:37 2010
New Revision: 168325

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=168325
Log:
    PR tree-optimization/47074
    * gimple-fold.c (gimplify_and_update_call_from_tree): Call
    pop_gimplify_context if returning early.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/gimple-fold.c

[Bug tree-optimization/47074] [4.6 Regression] tree check fail in create_access_replacement

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47074

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

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

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> 2010-12-30 08:26:17 UTC ---
Fixed.