[Bug middle-end/61919] New: [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[Bug middle-end/61919] New: [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

            Bug ID: 61919
           Summary: [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08   -O2
                     execution test
           Product: gcc
           Version: 4.10.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: schwab@linux-m68k.org
            Blocks: 61268
            Target: ia64-*-*

Created attachment 33187
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33187&action=edit
Output after r213002

The change for middle-end/61268 causes miscompilation of
gfortran.dg/fmt_g0_6.f08 on ia64.

$ gcc/gfortran -Bgcc/ -Bia64-suse-linux/./libgfortran/
../gcc/testsuite/gfortran.dg/fmt_g0_6.f08 -O2 -ffloat-store
-Bia64-suse-linux/./libgfortran/.libs -Lia64-suse-linux/./libgfortran/.libs
-Lia64-suse-linux/./libgfortran/.libs -Bia64-suse-linux/./libquadmath/.libs
-Lia64-suse-linux/./libquadmath/.libs -Lia64-suse-linux/./libquadmath/.libs -lm
-o ./fmt_g0_6.exe
$
LD_LIBRARY_PATH=ia64-suse-linux/libgfortran/.libs:ia64-suse-linux/libquadmath/.libs
./fmt_g0_6.exe 

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0  0x2000000000082E4F
#1  0x2000000000083F4F
#2  0xA0000000000407FF
#3  0x4000000000001D70 in check_f_fmt.854.constprop.1 at fmt_g0_6.f08:?
#4  0x4000000000001F9F in check_all.862.constprop.0 at fmt_g0_6.f08:?
Segmentation fault

[Bug middle-end/61919] [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

--- Comment #1 from Andreas Schwab <schwab@linux-m68k.org> ---
Created attachment 33188
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33188&action=edit
Output before r213002

[Bug middle-end/61919] [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2014-07-26
           Assignee|unassigned at gcc dot gnu.org      |rsandifo at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #2 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
Created attachment 33191
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33191&action=edit
Proposed patch

Looks like some invalid sharing from out-of-SSA.
Please try the attached patch.

[Bug middle-end/61919] [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

--- Comment #3 from Andreas Schwab <schwab@linux-m68k.org> ---
That fixes the test case.

[Bug middle-end/61919] [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.10.0

[Bug middle-end/61919] [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

--- Comment #4 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
Author: rsandifo
Date: Mon Jul 28 08:41:34 2014
New Revision: 213116

URL: https://gcc.gnu.org/viewcvs?rev=213116&root=gcc&view=rev
Log:
gcc/
    PR middle-end/61919
    * tree-outof-ssa.c (insert_partition_copy_on_edge)
    (insert_value_copy_on_edge, insert_rtx_to_part_on_edge)
    (insert_part_to_rtx_on_edge): Copy partition_to_pseudo rtxes before
    inserting them in the insn stream.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/tree-outof-ssa.c

[Bug middle-end/61919] [4.10 regression] FAIL: gfortran.dg/fmt_g0_6.f08 -O2 execution test

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61919

rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> changed:

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

--- Comment #5 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
Patch applied.