[Bug rtl-optimization/36438] New: gcc ICE compiling code with mmx builtin

[Bug rtl-optimization/36438] New: gcc ICE compiling code with mmx builtin

[davidxl at gcc dot gnu dot org]

Compiling the following code with latest compiler, got ice:

f.i: In function 'void foo(int __vector__*, int)':
f.i:33: internal compiler error: in trunc_int_for_mode, at explow.c:55
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.


// f.i:
typedef unsigned short int16;
typedef int __m64 __attribute__ ((__vector_size__ (8), __may_alias__));
typedef long long __v1di __attribute__ ((__vector_size__ (8)));

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__,
__artificial__))
_mm_slli_si64 (__m64 __m, int __count)
{
  return (__m64) __builtin_ia32_psllqi ((__v1di)__m, __count);
}

extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__,
__artificial__))
_mm_set_pi16 (short __w3, short __w2, short __w1, short __w0)
{
  return (__m64) __builtin_ia32_vec_init_v4hi (__w0, __w1, __w2, __w3);
}

inline __m64 __attribute__((__always_inline__)) SetS16(int16 a, int16 b, int16
c, int16 d) {
  return _mm_set_pi16(d, c, b, a);
}

void foo(__m64* dest, int n) {

  __m64 mask = SetS16(0x00FF, 0xFF00, 0x0000, 0x00FF);
  for ( int i = 0 ; i < n; ++i ) {

    mask = _mm_slli_si64(mask, 8);
    mask = _mm_slli_si64(mask, 8);

    *dest = mask;
    ++dest;
  }
  __builtin_ia32_emms ();
}


-- 
           Summary: gcc ICE compiling code with mmx builtin
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: rtl-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: davidxl at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[davidxl at gcc dot gnu dot org]

------- Comment #1 from davidxl at gcc dot gnu dot org  2008-06-05 06:41 -------

cse1 (RTL) does some expression simplification on the fly such as 

t = x << 4
r = t << 4

==>
r = x << 8

However for mmx shift operation, the mode (V1DI) for the const folding is
illegal -- resulting in ICE. 


-- 

davidxl at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |davidxl at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[ubizjak at gmail dot com]

------- Comment #2 from ubizjak at gmail dot com  2008-06-05 11:21 -------
Your testcase works for me on i686-pc-linux-gnu with 'GCC: (GNU) 4.4.0 20080605
(experimental) [trunk revision 136389]', with -mmmx and all optimization
levels:

.L5:
        movq    %mm1, %mm0
        psllq   $8, %mm0
        psllq   $8, %mm0
        movq    %mm0, (%ecx,%eax,8)
        addl    $1, %eax
        movq    %mm0, %mm1
        cmpl    %eax, %edx
        jg      .L5


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[ubizjak at gmail dot com]

------- Comment #3 from ubizjak at gmail dot com  2008-06-05 11:53 -------
The bug triggers on x86_64-pc-linux-gnu target. I will look into it.


-- 

ubizjak at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
 GCC target triplet|                            |x86_64-pc-linux-gnu
   Last reconfirmed|0000-00-00 00:00:00         |2008-06-05 11:53:55
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[ubizjak at gmail dot com]

------- Comment #4 from ubizjak at gmail dot com  2008-06-05 12:50 -------
There is a general problem with vector shifts by scalar operands. The code
assumed that both operands are vector mode operands, so it tries to add both
constant shift operands using V1DImode.

Following patch fixes ICE:

Index: cse.c
===================================================================
--- cse.c       (revision 136389)
+++ cse.c       (working copy)
@@ -3489,6 +3489,11 @@ fold_rtx (rtx x, rtx insn)
                          && exact_log2 (- INTVAL (const_arg1)) >= 0)))
                break;

+             /* ??? Vector mode shifts by scalar
+                shift operand are not supported yet.  */
+             if (is_shift && VECTOR_MODE_P (mode))
+               break;
+
              if (is_shift
                  && (INTVAL (inner_const) >= GET_MODE_BITSIZE (mode)
                      || INTVAL (inner_const) < 0))


-- 

ubizjak at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |ubizjak at gmail dot com
                   |dot org                     |
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2008-06-05 11:53:55         |2008-06-05 12:50:18
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[ubizjak at gmail dot com]

------- Comment #5 from ubizjak at gmail dot com  2008-06-05 17:05 -------
Patch at http://gcc.gnu.org/ml/gcc-patches/2008-06/msg00268.html


-- 

ubizjak at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2008-
                   |                            |06/msg00268.html


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[davidxl at gcc dot gnu dot org]

------- Comment #6 from davidxl at gcc dot gnu dot org  2008-06-05 17:37 -------
(In reply to comment #5)
> Patch at http://gcc.gnu.org/ml/gcc-patches/2008-06/msg00268.html
> 

Thanks -- same as my local workaround.

David


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[uros at gcc dot gnu dot org]

------- Comment #7 from uros at gcc dot gnu dot org  2008-06-06 15:06 -------
Subject: Bug 36438

Author: uros
Date: Fri Jun  6 15:04:51 2008
New Revision: 136486

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=136486
Log:
        PR rtl-optimization/36438
        * cse.c (fold_rtx) [ASHIFT, LSHIFTRT, ASHIFTRT]: Break out early
        for vector shifts with constant scalar shift operands.


Added:
    trunk/gcc/testsuite/gcc.target/i386/pr36438.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/cse.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[ubizjak at gmail dot com]

------- Comment #8 from ubizjak at gmail dot com  2008-06-06 15:08 -------
Fixed.


-- 

ubizjak at gmail dot com changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[uros at gcc dot gnu dot org]

------- Comment #9 from uros at gcc dot gnu dot org  2008-11-13 22:42 -------
Subject: Bug 36438

Author: uros
Date: Thu Nov 13 22:40:52 2008
New Revision: 141834

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141834
Log:
        Backport from mainline:
        2008-06-06  Uros Bizjak <ubizjak@gmail.com>

        PR rtl-optimization/36438
        * cse.c (fold_rtx) [ASHIFT, LSHIFTRT, ASHIFTRT]: Break out early
        for vector shifts with constant scalar shift operands.

testsuite/ChangeLog:

        Backport from mainline:
        2008-06-06  Uros Bizjak <ubizjak@gmail.com>

        PR rtl-optimization/36438
        * gcc.target/i386/pr36438.c


Added:
    branches/gcc-4_3-branch/gcc/testsuite/gcc.target/i386/pr36438.c
Modified:
    branches/gcc-4_3-branch/gcc/ChangeLog
    branches/gcc-4_3-branch/gcc/cse.c
    branches/gcc-4_3-branch/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438

[Bug rtl-optimization/36438] gcc ICE compiling code with mmx builtin

[uros at gcc dot gnu dot org]

------- Comment #10 from uros at gcc dot gnu dot org  2008-11-14 08:00 -------
Subject: Bug 36438

Author: uros
Date: Fri Nov 14 07:59:33 2008
New Revision: 141849

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141849
Log:
        Backport from mainline:
        2008-06-06  Uros Bizjak <ubizjak@gmail.com>

        PR rtl-optimization/36438
        * cse.c (fold_rtx) [ASHIFT, LSHIFTRT, ASHIFTRT]: Break out early
        for vector shifts with constant scalar shift operands.

testsuite/ChangeLog:

        Backport from mainline:
        2008-06-06  Uros Bizjak <ubizjak@gmail.com>

        PR rtl-optimization/36438
        * gcc.target/i386/pr36438.c


Added:
    branches/gcc-4_2-branch/gcc/testsuite/gcc.target/i386/pr36438.c
      - copied unchanged from r141834,
branches/gcc-4_3-branch/gcc/testsuite/gcc.target/i386/pr36438.c
Modified:
    branches/gcc-4_2-branch/gcc/ChangeLog
    branches/gcc-4_2-branch/gcc/cse.c
    branches/gcc-4_2-branch/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36438