[Bug tree-optimization/42640] New: wrong code for -ftree-loop-distribution in 175.vpr

[Bug tree-optimization/42640] New: wrong code for -ftree-loop-distribution in 175.vpr

[janis at gcc dot gnu dot org]

Benchmark test 175.vpr from SPEC CPU2000 gets incorrect results on
powerpc64-linux when compiled with "-O2 -ftree-loop-distribution", as shown by
a minimized testcase that I'll attach to this PR.  The inner loop is:

      inode = s_node;
      for (iloop = 1; iloop <= 2; iloop++)
        {
          rr_node[inode].a = i;
          rr_node[inode].b = j;
          rr_node[inode].c = ipad;

          inode = p_node;
        }

Expected output:

elm3b149% /home/janis/tools/gcc-trunk-anonsvn-32/bin/gcc -O2 bug.c && ./a.out
rr_node[0] = 1 2 0
rr_node[1] = 1 2 1
rr_node[2] = 1 2 0
rr_node[3] = 1 2 1

Bad output:

elm3b149% /home/janis/tools/gcc-trunk-anonsvn-32/bin/gcc -O2
-ftree-loop-distribution bug.c && ./a.out
rr_node[0] = 1 2 0
rr_node[1] = 1 2 1
rr_node[2] = 0 0 0
rr_node[3] = 0 0 1

The test also fails with GCC 4.4.2.


-- 
           Summary: wrong code for -ftree-loop-distribution in 175.vpr
           Product: gcc
           Version: 4.5.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc64-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[janis at gcc dot gnu dot org]

------- Comment #1 from janis at gcc dot gnu dot org  2010-01-06 23:11 -------
Created an attachment (id=19493)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=19493&action=view)
minimized executable testcase


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[zsojka at seznam dot cz]

------- Comment #2 from zsojka at seznam dot cz  2010-02-08 15:28 -------
I can reproduce this on x86_64, trunk r156598 and 4.4 r156256.

Also, bootstrap fails for me with BOOT_CFLAGS="-O2 -ftree-loop-distribution":
Comparing stages 2 and 3
warning: gcc/cc1plus-checksum.o differs
warning: gcc/cc1-checksum.o differs
Bootstrap comparison failure!
libiberty/sha1.o differs
make[2]: *** [compare] Error 1


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[janis at gcc dot gnu dot org]

------- Comment #3 from janis at gcc dot gnu dot org  2010-02-23 19:27 -------
The minimized testcase and 175.vpr still fail on powerpc64-linux with "-O2
-ftree-loop-distribution" for both -m32 and -m64.  With the confirmation in
comment #2 this report ought to be moved from UNCONFIRMED to NEW.

This bug is not a regression from 4.4 but it is a wrong-code bug and ought to
be considered for fixing in 4.5.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[bangerth at gmail dot com]

-- 

bangerth at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mmitchel at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2010-02-23 20:38:36
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[spop at gcc dot gnu dot org]

------- Comment #4 from spop at gcc dot gnu dot org  2010-02-23 21:07 -------
Reza Yazdani has a patch to fix this.
He's going to post the patch after test.

Sebastian


-- 

spop at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |reza dot yazdani at amd dot
                   |dot org                     |com
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[spop at gcc dot gnu dot org]

------- Comment #5 from spop at gcc dot gnu dot org  2010-03-02 10:23 -------
Subject: Bug 42640

Author: spop
Date: Tue Mar  2 10:22:30 2010
New Revision: 157161

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=157161
Log:
Fix PR42640: Correctly initialize the value of the new induction variable.

2010-03-02  Reza Yazdani  <reza.yazdani@amd.com>

        PR middle-end/42640
        * tree-loop-distribution.c (update_phis_for_loop_copy): Replaced
        the assignment from the new induction variable to the assignment
        of the value from the original loop PHI function.

        * gcc.dg/tree-ssa/pr42640.c: New.

Added:
    trunk/gcc/testsuite/gcc.dg/tree-ssa/pr42640.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-loop-distribution.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640

[Bug tree-optimization/42640] wrong code for -ftree-loop-distribution in 175.vpr

[spop at gcc dot gnu dot org]

------- Comment #6 from spop at gcc dot gnu dot org  2010-03-02 10:24 -------
Fixed.


-- 

spop at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42640