[Bug rtl-optimization/34283] New: Non-optimal reload register used

[Bug rtl-optimization/34283] New: Non-optimal reload register used

[ubizjak at gmail dot com]

Following testcase:

--cut here--
typedef long long __m128i __attribute__ ((__vector_size__ (16),
__may_alias__));
typedef long long __v2di __attribute__ ((__vector_size__ (16)));


__m128i _mm_set_epi64x (long long __q1, long long __q0)
{
  return __extension__ (__m128i)(__v2di){ __q0, __q1 };
}
--cut here--

compiles using -O2 -msse2 into:

_mm_set_epi64x:
        movq    12(%esp), %xmm1
        movhps  4(%esp), %xmm1
(1)     movdqa  %xmm1, %xmm0
        ret

insn (1) is not needed.

Before reload, we have:

(insn:HI 12 7 18 2 epi.c:8 (set (reg/i:V2DI 21 xmm0 [ <result> ])
        (vec_concat:V2DI (mem/c/i:DI (plus:SI (reg/f:SI 16 argp)
                    (const_int 8 [0x8])) [2 __q0+0 S8 A32])
            (mem/c/i:DI (reg/f:SI 16 argp) [2 __q1+0 S8 A32]))) 1039
{vec_concatv2di} (nil))

(insn:HI 18 12 0 2 epi.c:8 (use (reg/i:V2DI 21 xmm0 [ <result> ])) -1 (nil))


For some reason reload isn't satisfied with %xmm0:

Reload 0: reload_in (DI) = (mem/c/i:DI (plus:SI (reg/f:SI 7 sp)
                                                        (const_int 12 [0xc]))
[2 __q0+0 S8 A32])
        reload_out (V2DI) = (reg/i:V2DI 21 xmm0 [ <result> ])
        SSE_REGS, RELOAD_OTHER (opnum = 0)
        reload_in_reg: (mem/c/i:DI (plus:SI (reg/f:SI 7 sp)
                                                        (const_int 12 [0xc]))
[2 __q0+0 S8 A32])
        reload_out_reg: (reg/i:V2DI 21 xmm0 [ <result> ])
        reload_reg_rtx: (reg:V2DI 22 xmm1)


The insn pattern is:

  [(set (match_operand:V2DI 0 "register_operand"     "=Y2,?Y2,Y2,x,x,x")
        (vec_concat:V2DI
          (match_operand:DI 1 "nonimmediate_operand" "  m,*y ,0 ,0,0,m")
          (match_operand:DI 2 "vector_move_operand"  "  C,  C,Y2,x,m,0")))]

So, reload should take into account that operand 1 should match operand 0, so
%xmm0 should be used.


-- 
           Summary: Non-optimal reload register used
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: rtl-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ubizjak at gmail dot com
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

[Bug rtl-optimization/34283] Non-optimal reload register used

[steven at gcc dot gnu dot org]

------- Comment #1 from steven at gcc dot gnu dot org  2009-06-22 22:18 -------
Uros, this bug is from the pre-IRA times.  Could you check if you still see
this problem, please?


-- 

steven at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

[Bug rtl-optimization/34283] Non-optimal reload register used

[ubizjak at gmail dot com]

------- Comment #2 from ubizjak at gmail dot com  2009-06-23 12:09 -------
(In reply to comment #1)
> Uros, this bug is from the pre-IRA times.  Could you check if you still see
> this problem, please?

Yes, it is still present as of gcc 4.5.0 20090623


-- 

ubizjak at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|0000-00-00 00:00:00         |2009-06-23 12:09:16
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

[Bug rtl-optimization/34283] Non-optimal reload register used

[ubizjak at gmail dot com]

------- Comment #3 from ubizjak at gmail dot com  2009-10-03 08:36 -------
(In reply to comment #2)

> > Uros, this bug is from the pre-IRA times.  Could you check if you still see
> > this problem, please?
> 
> Yes, it is still present as of gcc 4.5.0 20090623

Reconfirmed with gcc version 4.5.0 20091003 (experimental) [trunk revision
152431].


-- 

ubizjak at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|2009-06-23 12:09:16         |2009-10-03 08:36:28
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

[Bug rtl-optimization/34283] Non-optimal reload register used

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

Uros Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.8.0

--- Comment #8 from Uros Bizjak <ubizjak at gmail dot com> 2012-10-27 07:53:30 UTC ---
Fixed.

[Bug rtl-optimization/34283] Non-optimal reload register used

[uros at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

--- Comment #7 from uros at gcc dot gnu.org 2012-10-27 07:49:00 UTC ---
Author: uros
Date: Sat Oct 27 07:48:55 2012
New Revision: 192872

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=192872
Log:
    PR target/34283
    * gcc.target/i386/pr34283.c: New test.


Added:
    trunk/gcc/testsuite/gcc.target/i386/pr34283.c
Modified:
    trunk/gcc/testsuite/ChangeLog

[Bug rtl-optimization/34283] Non-optimal reload register used

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

--- Comment #6 from Uros Bizjak <ubizjak at gmail dot com> 2012-10-27 07:33:41 UTC ---
(In reply to comment #5)
> (In reply to comment #4)
> > Still present on x86_64. -O2 -march=corei7:
> > 
> >         movd    %rsi, %xmm1
> >         pinsrq  $1, %rdi, %xmm1
> >         movdqa  %xmm1, %xmm0
> >         ret
> 
> It was still like that yesterday, but today:
> 
>     movd    %rsi, %xmm0
>     pinsrq    $1, %rdi, %xmm0
>     ret
> 
> Fixed by LRA?

Looks so. I will prepare a testcase and close the bug.

[Bug rtl-optimization/34283] Non-optimal reload register used

[glisse at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |glisse at gcc dot gnu.org

--- Comment #5 from Marc Glisse <glisse at gcc dot gnu.org> 2012-10-23 18:33:07 UTC ---
(In reply to comment #4)
> Still present on x86_64. -O2 -march=corei7:
> 
>         movd    %rsi, %xmm1
>         pinsrq  $1, %rdi, %xmm1
>         movdqa  %xmm1, %xmm0
>         ret

It was still like that yesterday, but today:

    movd    %rsi, %xmm0
    pinsrq    $1, %rdi, %xmm0
    ret

Fixed by LRA?

[Bug rtl-optimization/34283] Non-optimal reload register used

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34283

Uros Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |vmakarov at gcc dot gnu.org
      Known to fail|                            |4.7.0

--- Comment #4 from Uros Bizjak <ubizjak at gmail dot com> 2012-01-20 10:15:32 UTC ---
Still present on x86_64. -O2 -march=corei7:

        movd    %rsi, %xmm1
        pinsrq  $1, %rdi, %xmm1
        movdqa  %xmm1, %xmm0
        ret

GCC: (GNU) 4.7.0 20120118 (experimental) [trunk revision 183277]

Re-confirmed and added CC.