[Bug tree-optimization/48498] New: Several gcc.dg/vect tests XPASS on SPARC

[Bug tree-optimization/48498] New: Several gcc.dg/vect tests XPASS on SPARC

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

           Summary: Several gcc.dg/vect tests XPASS on SPARC
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: ro@gcc.gnu.org
                CC: ebotcazou@gcc.gnu.org
              Host: sparc*-*-*
            Target: sparc*-*-*
             Build: sparc*-*-*


Created attachment 23915
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23915
slp-3.c -fdump-tree-vect-details  output

Several gcc.dg/vect tests started to XPASS on SPARC since around 20110331:

XPASS: gcc.dg/vect/slp-3.c scan-tree-dump-times vect "vectorized 3 loops" 1
XPASS: gcc.dg/vect/slp-3.c scan-tree-dump-times vect "vectorizing stmts using
SLP" 3
XPASS: gcc.dg/vect/no-vfa-pr29145.c scan-tree-dump-times vect "vectorized 0
loops" 2
XPASS: gcc.dg/vect/no-vfa-pr29145.c scan-tree-dump-times vect "vectorized 1
loops" 1

I'm attaching the vect dumps.

[Bug tree-optimization/48498] Several gcc.dg/vect tests XPASS on SPARC

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #1 from Rainer Orth <ro at gcc dot gnu.org> 2011-04-07 17:38:05 UTC ---
Created attachment 23916
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23916
no-vfa-pr29145.c -fdump-tree-vect-details output

[Bug tree-optimization/48498] Several gcc.dg/vect tests XPASS on SPARC

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

Ira Rosen <irar at il dot ibm.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |irar at il dot ibm.com

--- Comment #2 from Ira Rosen <irar at il dot ibm.com> 2011-04-17 12:15:29 UTC ---
Could you please attach vect dumps before the XPASSes? From the current dumps I
can only understand that unaligned accesses get vectorized by forcing alignment
of local arrays.

Thanks,
Ira

[Bug tree-optimization/48498] Several gcc.dg/vect tests XPASS on SPARC

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #3 from Rainer Orth <ro at gcc dot gnu.org> 2011-04-18 14:35:21 UTC ---
Created attachment 24032
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=24032
slp-3.c -fdump-tree-vect-details  output, 4.6 branch

[Bug tree-optimization/48498] Several gcc.dg/vect tests XPASS on SPARC

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #4 from Rainer Orth <ro at gcc dot gnu.org> 2011-04-18 14:36:09 UTC ---
Created attachment 24033
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=24033
no-vfa-pr29145.c -fdump-tree-vect-details output, 4.6 branch

[Bug tree-optimization/48498] Several gcc.dg/vect tests XPASS on SPARC

[ro at CeBiTec dot Uni-Bielefeld.DE]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #5 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> 2011-04-18 14:37:18 UTC ---
> --- Comment #2 from Ira Rosen <irar at il dot ibm.com> 2011-04-17 12:15:29 UTC ---
> Could you please attach vect dumps before the XPASSes? From the current dumps I
> can only understand that unaligned accesses get vectorized by forcing alignment
> of local arrays.

Sure, I've just attached the corresponding dumps from the 4.6 branch
that still XFAIL.

    Rainer

[Bug tree-optimization/48498] Several gcc.dg/vect tests XPASS on SPARC

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #6 from Ira Rosen <irar at il dot ibm.com> 2011-04-20 09:38:20 UTC ---
Thanks Rainer.

It is caused by my patch
http://gcc.gnu.org/viewcvs?view=revision&revision=171569 that changed these
tests to work correctly on NEON doubleword vectors. The patch actually fixed
these tests for SPARC as well, because SPARC also uses smaller vector size. The
problem is that these tests are expected to fail on vect_no_align targets, but
actually no realignment mechanism is needed here except for forcing alignment
of arrays or having them aligned. We don't have a keyword for this and I think
it is hard to have such because it's a combination of the target behavior and
the test.

We can try to remove vect_no_align keyword from these tests and check what
happens on other no_align targets
        if { [istarget mipsisa64*-*-*]
             || [istarget sparc*-*-*]
             || [istarget ia64-*-*]
             || [check_effective_target_arm_vect_no_misalign]
             || ([istarget mips*-*-*]
                 && [check_effective_target_mips_loongson]) } {
            set et_vect_no_align_saved 1

I checked it with a cross-compiler on ia64 (which I think is supposed to have a
similar behavior to SPARC here). I think the tests XPASS on ia64 as well at the
moment.

I'll submit a patch with RFT.

[Bug testsuite/48498] Several gcc.dg/vect tests XPASS on SPARC

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #7 from Ira Rosen <irar at il dot ibm.com> 2011-04-28 07:42:31 UTC ---
I submitted a patch (http://gcc.gnu.org/ml/gcc-patches/2011-04/msg01627.html)
some time ago, but there are no volunteers to test it so far. 

Rainer, could you please check it on SPARC?

Thanks,
Ira

[Bug testsuite/48498] Several gcc.dg/vect tests XPASS on SPARC

[ro at CeBiTec dot Uni-Bielefeld.DE]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #8 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> 2011-04-29 10:42:00 UTC ---
> --- Comment #7 from Ira Rosen <irar at il dot ibm.com> 2011-04-28 07:42:31 UTC ---
> I submitted a patch (http://gcc.gnu.org/ml/gcc-patches/2011-04/msg01627.html)
> some time ago, but there are no volunteers to test it so far. 

I think I even noticed it, but somehow if fell through the cracks.

> Rainer, could you please check it on SPARC?

I had to apply it manually (it seems your mailer mangled it badly), but
it fixed the remaining XPASSes in a sparc-sun-solaris2.11 bootstrap.

Thanks.
        Rainer

[Bug testsuite/48498] Several gcc.dg/vect tests XPASS on SPARC

[ebotcazou at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011.05.02 07:30:38
     Ever Confirmed|0                           |1

--- Comment #9 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2011-05-02 07:30:38 UTC ---
Yep.

[Bug testsuite/48498] Several gcc.dg/vect tests XPASS on SPARC

[irar at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

--- Comment #10 from irar at gcc dot gnu.org 2011-05-02 11:13:31 UTC ---
Author: irar
Date: Mon May  2 11:13:26 2011
New Revision: 173245

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=173245
Log:

    PR testsuite/48498
    * gcc.dg/vect/slp-3.c: Increase loop bound.  Don't expect to fail
    on vect_no_align targets.
    * gcc.dg/vect/no-vfa-pr29145.c: Don't expect to fail on
    vect_no_align targets.


Modified:
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gcc.dg/vect/no-vfa-pr29145.c
    trunk/gcc/testsuite/gcc.dg/vect/slp-3.c

[Bug testsuite/48498] Several gcc.dg/vect tests XPASS on SPARC

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48498

Ira Rosen <irar at il dot ibm.com> changed:

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

--- Comment #11 from Ira Rosen <irar at il dot ibm.com> 2011-05-02 11:14:11 UTC ---
Fixed.