[Bug ipa/66863] New: wrong code at -Os and above on x86_64-linux-gnu

[Bug ipa/66863] New: wrong code at -Os and above on x86_64-linux-gnu

[su at cs dot ucdavis.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66863

            Bug ID: 66863
           Summary: wrong code at -Os and above on x86_64-linux-gnu
           Product: gcc
           Version: 6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: ipa
          Assignee: unassigned at gcc dot gnu.org
          Reporter: su at cs dot ucdavis.edu
  Target Milestone: ---

The current gcc trunk mis-compiles the following code on x86_64-linux-gnu at
-Os and above in both 32-bit and 64-bit modes. 

This is a regression from 5.1.x.

$ gcc-trunk -v
Using built-in specs.
COLLECT_GCC=gcc-trunk
COLLECT_LTO_WRAPPER=/usr/local/gcc-trunk/libexec/gcc/x86_64-unknown-linux-gnu/6.0.0/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../gcc-trunk/configure --prefix=/usr/local/gcc-trunk
--enable-languages=c,c++ --disable-werror --enable-multilib
Thread model: posix
gcc version 6.0.0 20150713 (experimental) [trunk revision 225727] (GCC) 
$ 
$ gcc-trunk -O1 small.c; ./a.out
$ gcc-5.1 -Os small.c; ./a.out
$ 
$ gcc-trunk -Os small.c
$ ./a.out
Floating point exception (core dumped)
$ 


---------------------------------


int a, b;

int
fn1 (int p1)
{
  if (p1 < -2147483647) 
    return 0;
  else 
    return 1;
}

int
fn2 (int p1, short p2)
{
  return p2 ? p1 % p2 : 0; 
}

int
main ()
{
  b = fn2 (fn1 (a), a);
  return 0;
}

[Bug tree-optimization/66863] [6 Regression] wrong code at -Os and above on x86_64-linux-gnu

[glisse at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66863

Marc Glisse <glisse at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-07-14
          Component|ipa                         |tree-optimization
   Target Milestone|---                         |6.0
            Summary|wrong code at -Os and above |[6 Regression] wrong code
                   |on x86_64-linux-gnu         |at -Os and above on
                   |                            |x86_64-linux-gnu
     Ever confirmed|0                           |1

--- Comment #1 from Marc Glisse <glisse at gcc dot gnu.org> ---
In VRP:

  _3 = (short int) a.0_2;
  if (a.0_2 == -2147483648)
    goto <bb 7>;
  else
    goto <bb 3>;

  <bb 7>:
  _10 = ASSERT_EXPR <_3, _3 == 0>;
  goto <bb 4>;

  <bb 3>:
  _11 = ASSERT_EXPR <_3, _3 != 0>;

The == ASSERT_EXPR looks fine, but this last != ASSERT_EXPR is a bit
optimistic.

[Bug tree-optimization/66863] [6 Regression] wrong code at -Os and above on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66863

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Probably caused by

2015-06-30  Richard Biener  <rguenther@suse.de>

        * tree-vrp.c (register_edge_assert_for_2): Also register
        asserts for dominating conversion results.

I'll have a look.

[Bug tree-optimization/66863] [6 Regression] wrong code at -Os and above on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66863

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Tue Jul 14 11:39:14 2015
New Revision: 225768

URL: https://gcc.gnu.org/viewcvs?rev=225768&root=gcc&view=rev
Log:
2015-07-14  Richard Biener  <rguenther@suse.de>

        PR tree-optimization/66863
        * tree-vrp.c (register_edge_assert_for_2): Properly restrict
        what we record for conversion use stmt lhs inequalities.

        * gcc.dg/torture/pr66863.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/torture/pr66863.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-vrp.c

[Bug tree-optimization/66863] [6 Regression] wrong code at -Os and above on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66863

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

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

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.