[Bug middle-end/49840] New: [4.7 Regression] New test failures

[Bug middle-end/49840] New: [4.7 Regression] New test failures

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

           Summary: [4.7 Regression] New test failures
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: hjl.tools@gmail.com
                CC: rguenth@gcc.gnu.org


On Linux/ia32, revision 176737 gave:

FAIL: c-c++-common/dfp/usual-arith-conv.c execution test
FAIL: c-c++-common/dfp/usual-arith-conv.c execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O0  execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O1  execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O2  execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O2 -flto  execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O2 -flto -flto-partition=none 
execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O3 -fomit-frame-pointer 
execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -O3 -g  execution test
FAIL: gcc.dg/torture/fp-int-convert-float80.c  -Os  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O0  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O1  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O2  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O2 -flto  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O2 -flto
-flto-partition=none  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O3 -fomit-frame-pointer 
execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -O3 -g  execution test
FAIL: gcc.dg/torture/fp-int-convert-long-double.c  -Os  execution test
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O0 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O1 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O2 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O2
-fbounds-check 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O2
-fomit-frame-pointer -finline-functions 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O2
-fomit-frame-pointer -finline-functions -funroll-loops 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -O3 -g 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution,  -Os 
FAIL: gfortran.fortran-torture/execute/random_2.f90 execution, -O2
-ftree-vectorize -msse2 

Revision 176734 is OK.

[Bug middle-end/49840] [4.7 Regression] New test failures

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |i?86-*-*
   Target Milestone|---                         |4.7.0

--- Comment #1 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-07-25 14:17:12 UTC ---
Can't reproduce this on a x86_64 host.  I tried -m32 -mno-sse -march=i386
and also -march=i686.

[Bug middle-end/49840] [4.7 Regression] New test failures

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-07-25 14:27:50 UTC ---
Does

Index: gcc/tree-vrp.c
===================================================================
--- gcc/tree-vrp.c      (revision 176745)
+++ gcc/tree-vrp.c      (working copy)
@@ -7432,11 +7465,15 @@ range_fits_type_p (value_range_t *vr, un
     return false;

   tem = double_int_ext (tree_to_double_int (vr->min), precision, unsigned_p);
-  if (!double_int_equal_p (tree_to_double_int (vr->min), tem))
+  if ((TYPE_UNSIGNED (TREE_TYPE (vr->min)) != unsigned_p
+       && (TREE_INT_CST_HIGH (vr->min) < 0) != (tem.high < 0))
+      || !double_int_equal_p (tree_to_double_int (vr->min), tem))
     return false;

   tem = double_int_ext (tree_to_double_int (vr->max), precision, unsigned_p);
-  if (!double_int_equal_p (tree_to_double_int (vr->max), tem))
+  if ((TYPE_UNSIGNED (TREE_TYPE (vr->max)) != unsigned_p
+       && (TREE_INT_CST_HIGH (vr->max) < 0) != (tem.high < 0))
+      || !double_int_equal_p (tree_to_double_int (vr->max), tem))
     return false;

   return true;

fix it?

[Bug middle-end/49840] [4.7 Regression] New test failures

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.7.0                       |---

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> 2011-07-25 14:30:49 UTC ---
(In reply to comment #1)
> Can't reproduce this on a x86_64 host.  I tried -m32 -mno-sse -march=i386
> and also -march=i686.

On Linux/x86-64 host, please configure gcc with

CC="gcc -m32" CXX="g++ -m32" ../src-trunk/configure i686-linux

This is how I test GCC for Linux/ia32 on Linux/x86-64.

[Bug middle-end/49840] [4.7 Regression] New test failures

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2011.07.26 08:50:04
         AssignedTo|unassigned at gcc dot       |rguenth at gcc dot gnu.org
                   |gnu.org                     |
   Target Milestone|---                         |4.7.0
     Ever Confirmed|0                           |1

--- Comment #4 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-07-26 08:50:04 UTC ---
It looks like __fixunsxfdi is miscompiled.

[Bug middle-end/49840] [4.7 Regression] New test failures

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

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> 2011-07-26 12:37:20 UTC ---
Fixed.

[Bug middle-end/49840] [4.7 Regression] New test failures

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49840

--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-07-26 12:37:03 UTC ---
Author: rguenth
Date: Tue Jul 26 12:37:00 2011
New Revision: 176790

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=176790
Log:
2011-07-26  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/49840
    * tree-vrp.c (range_fits_type_p): Properly handle full
    double-int precision.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/tree-vrp.c