[Bug tree-optimization/47411] New: [4.5 Regression] Bootstrap failure on x86-64/Darwin

[Bug tree-optimization/47411] New: [4.5 Regression] Bootstrap failure on x86-64/Darwin

[ebotcazou at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

           Summary: [4.5 Regression] Bootstrap failure on x86-64/Darwin
           Product: gcc
           Version: 4.5.3
            Status: UNCONFIRMED
          Severity: critical
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: ebotcazou@gcc.gnu.org


The backport

2011-01-17  Richard Guenther  <rguenther@suse.de>

        PR tree-optimization/44592
        * tree-ssa-ccp.c (gimplify_and_update_call_from_tree): Copy
        from trunk.

has introduced a bootstrap failure for x86-64/Darwin on the 4.5 branch:

/lena.a/gnatmail/gcc-45/build-lena/obj/./prev-gcc/xgcc
-B/lena.a/gnatmail/gcc-45/build-lena/obj/./prev-gcc/
-B/usr/local/gnat/x86_64-apple-darwin10.2.0/bin/
-B/usr/local/gnat/x86_64-apple-darwin10.2.0/bin/
-B/usr/local/gnat/x86_64-apple-darwin10.2.0/lib/ -isystem
/usr/local/gnat/x86_64-apple-darwin10.2.0/include -isystem
/usr/local/gnat/x86_64-apple-darwin10.2.0/sys-include    -c  -O2 -gtoggle
-DIN_GCC   -W -Wall -Wwrite-strings -Wcast-qual -Wstrict-prototypes
-Wmissing-prototypes -Wmissing-format-attribute -pedantic -Wno-long-long
-Wno-variadic-macros -Wno-overlength-strings -Wold-style-definition
-Wc++-compat   -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild -I../../src/gcc
-I../../src/gcc/build -I../../src/gcc/../include
-I../../src/gcc/../libcpp/include
-I/lena.a/gnatmail/gcc-45/build-lena/libmpfr/install/include
-I/lena.a/gnatmail/gcc-45/build-lena/libmpfr/install/include
-I/lena.a/gnatmail/gcc-45/build-lena/libmpfr/install/include 
-I../../src/gcc/../libdecnumber -I../../s
 rc/gcc/../libdecnumber/dpd -I../libdecnumber     \
                -o build/gengenrtl.o ../../src/gcc/gengenrtl.c
../../src/gcc/genmodes.c: In function 'new_mode':
../../src/gcc/genmodes.c:153:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.

Reduced testcase to be attached for x86/Linux.  The problem is that a new SSA
name is created in gimplify_and_update_call_from_tree:

Program received signal SIGSEGV, Segmentation fault.
0x085a0139 in vrp_finalize ()
    at /home/eric/gnat/gnat6_45/src/gcc/tree-vrp.c:7338
7338            BITMAP_FREE (vr_value[i]->equiv);
(gdb) p cfun->gimple_df->ssa_names.base
$12 = {num = 22, alloc = 50, vec = {0x0}}

Breakpoint 1, vrp_initialize ()
    at /home/eric/gnat/gnat6_45/src/gcc/tree-vrp.c:5354
5354      vr_value = XCNEWVEC (value_range_t *, num_ssa_names);
(gdb) p cfun->gimple_df->ssa_names.base
$13 = {num = 21, alloc = 50, vec = {0x0}}

Hardware watchpoint 2: { unsigned int } 0xf7ca5a00

Old value = 21
New value = 22
0x0855e711 in VEC_tree_base_quick_push (
    function_=0x8b7a17e "make_ssa_name_fn", line_=146, obj_=0xf7d26d20,
    vec_=0xf7ca5a00,
    file_=0x8b7a0e8 "/home/eric/gnat/gnat6_45/src/gcc/tree-ssanames.c")
    at /home/eric/gnat/gnat6_45/src/gcc/tree.h:182
182     DEF_VEC_P(tree);
(gdb) bt
#0  0x0855e711 in VEC_tree_base_quick_push (
    function_=0x8b7a17e "make_ssa_name_fn", line_=146, obj_=0xf7d26d20,
    vec_=0xf7ca5a00,
    file_=0x8b7a0e8 "/home/eric/gnat/gnat6_45/src/gcc/tree-ssanames.c")
    at /home/eric/gnat/gnat6_45/src/gcc/tree.h:182
#1  VEC_tree_gc_safe_push (function_=0x8b7a17e "make_ssa_name_fn", line_=146,
    file_=0x8b7a0e8 "/home/eric/gnat/gnat6_45/src/gcc/tree-ssanames.c",
    obj_=0xf7d26d20, vec_=0xf7cae918)
    at /home/eric/gnat/gnat6_45/src/gcc/tree.h:183
#2  make_ssa_name_fn (fn=0xf7cb2210, var=0xf7d29564, stmt=0xf7d26cc0)
    at /home/eric/gnat/gnat6_45/src/gcc/tree-ssanames.c:146
#3  0x084dbcf0 in make_ssa_name (stmt=0xf7d26cc0, var=<value optimized out>)
    at /home/eric/gnat/gnat6_45/src/gcc/tree-flow-inline.h:1247
#4  gimplify_and_update_call_from_tree (si_p=0xffffcd68,
    expr=<value optimized out>)
    at /home/eric/gnat/gnat6_45/src/gcc/tree-ssa-ccp.c:3434

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[ebotcazou at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

--- Comment #1 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2011-01-22 09:41:19 UTC ---
Created attachment 23075
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23075
Reduced testcase for x86

To be compiled at -O2.

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[bosch at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

Geert Bosch <bosch at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011.01.22 12:08:12
                 CC|                            |bosch at gcc dot gnu.org
     Ever Confirmed|0                           |1

--- Comment #2 from Geert Bosch <bosch at gcc dot gnu.org> 2011-01-22 12:08:12 UTC ---
I've confirmed that reverting Richard's patch to tree-ssa-ccp.c restores
bootstrap.

  -Geert

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1
             Status|NEW                         |ASSIGNED
         AssignedTo|rguenther at suse dot de    |rguenth at gcc dot gnu.org
   Target Milestone|---                         |4.5.3

--- Comment #3 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-24 09:43:17 UTC ---
Mine.

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

--- Comment #4 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-24 16:53:07 UTC ---
I think a backport of

2010-06-30  Michael Matz  <matz@suse.de>

        PR bootstrap/44699
        * tree-vrp.c (vrp_finalize): Deal with changing num_ssa_names.
        * gimple-fold.c (gimplify_and_update_call_from_tree): If LHS is
        a gimple reg, attach the original VDEF to the last store in the
        sequence.

is missing.

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-25 09:48:12 UTC ---
Author: rguenth
Date: Tue Jan 25 09:48:07 2011
New Revision: 169222

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=169222
Log:
2011-01-25  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/47411
    Backport from mainline
    2010-06-30  Michael Matz  <matz@suse.de>

    PR bootstrap/44699
    * tree-vrp.c (vrp_finalize): Deal with changing num_ssa_names.

Modified:
    branches/gcc-4_5-branch/gcc/ChangeLog
    branches/gcc-4_5-branch/gcc/tree-vrp.c

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

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

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

--- Comment #6 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-25 09:49:29 UTC ---
Should be fixed.

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

--- Comment #8 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-25 10:26:23 UTC ---
Author: rguenth
Date: Tue Jan 25 10:26:20 2011
New Revision: 169225

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=169225
Log:
2011-01-25  Richard Guenther  <rguenther@suse.de>

    PR middle-end/47411
    * gcc.dg/torture/pr47411.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/torture/pr47411.c
Modified:
    trunk/gcc/testsuite/ChangeLog

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

--- Comment #7 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-01-25 10:24:59 UTC ---
Author: rguenth
Date: Tue Jan 25 10:24:56 2011
New Revision: 169224

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=169224
Log:
2011-01-25  Richard Guenther  <rguenther@suse.de>

    PR middle-end/47411
    * gcc.dg/torture/pr47411.c: New testcase.

Added:
    branches/gcc-4_5-branch/gcc/testsuite/gcc.dg/torture/pr47411.c
Modified:
    branches/gcc-4_5-branch/gcc/testsuite/ChangeLog

[Bug tree-optimization/47411] [4.5 Regression] Bootstrap failure on x86-64/Darwin

[dnovillo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47411

--- Comment #9 from Diego Novillo <dnovillo at gcc dot gnu.org> 2011-02-02 17:46:00 UTC ---
Author: dnovillo
Date: Wed Feb  2 17:45:57 2011
New Revision: 169576

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=169576
Log:
2011-01-25  Richard Guenther  <rguenther@suse.de>

    PR middle-end/47411
    * gcc.dg/torture/pr47411.c: New testcase.

Added:
    branches/google/integration/gcc/testsuite/gcc.dg/torture/pr47411.c
Modified:
    branches/google/integration/gcc/testsuite/ChangeLog