[Bug rtl-optimization/47299] New: Widening multiply optimization generates bad code

[Bug rtl-optimization/47299] New: Widening multiply optimization generates bad code

[uweigand at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

           Summary: Widening multiply optimization generates bad code
           Product: gcc
           Version: 4.5.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: rtl-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: uweigand@gcc.gnu.org
                CC: bernds@codesourcery.com, rguenther@suse.de


Building the following test case with current mainline on i386:

unsigned short test (unsigned char val) __attribute__ ((noinline));

unsigned short
test (unsigned char val)
{
  return val * 255;
}

int
main(int argc, char**argv)
{
  printf ("test(val=40) = %x\n", test(0x40));
  return 0;
}

We get the following (correct) output with -O0:
test(val=40) = 3fc0

and the following incorrect output with -O2:
test(val=40) = ffc0

The problem appears to be related to this piece of code in expand_expr_real2,
case WIDEN_MULT_EXPR:

                  expand_operands (treeop0, treeop1, NULL_RTX, &op0, &op1,
                                   EXPAND_NORMAL);
                  temp = expand_widening_mult (mode, op0, op1, target,
                                               unsignedp, this_optab);

expand_operands will expand the constant 255 into QImode and return a
(const_int -1) for op1.  Passing this constant into expand_widening_mult then
apparently generates a simple negation operation in HImode instead (via
expand_const_mult) ...

It seems this code came in here:
http://gcc.gnu.org/ml/gcc-patches/2010-04/msg01327.html
Any suggestions how this ought to be handled?

[Bug rtl-optimization/47299] Widening multiply optimization generates bad code

[rguenther at suse dot de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

--- Comment #1 from rguenther at suse dot de <rguenther at suse dot de> 2011-01-14 21:49:32 UTC ---
On Fri, 14 Jan 2011, uweigand at gcc dot gnu.org wrote:

> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299
> 
>            Summary: Widening multiply optimization generates bad code
>            Product: gcc
>            Version: 4.5.0
>             Status: UNCONFIRMED
>           Keywords: wrong-code
>           Severity: normal
>           Priority: P3
>          Component: rtl-optimization
>         AssignedTo: unassigned@gcc.gnu.org
>         ReportedBy: uweigand@gcc.gnu.org
>                 CC: bernds@codesourcery.com, rguenther@suse.de
> 
> 
> Building the following test case with current mainline on i386:
> 
> unsigned short test (unsigned char val) __attribute__ ((noinline));
> 
> unsigned short
> test (unsigned char val)
> {
>   return val * 255;
> }
> 
> int
> main(int argc, char**argv)
> {
>   printf ("test(val=40) = %x\n", test(0x40));
>   return 0;
> }
> 
> We get the following (correct) output with -O0:
> test(val=40) = 3fc0
> 
> and the following incorrect output with -O2:
> test(val=40) = ffc0
> 
> The problem appears to be related to this piece of code in expand_expr_real2,
> case WIDEN_MULT_EXPR:
> 
>                   expand_operands (treeop0, treeop1, NULL_RTX, &op0, &op1,
>                                    EXPAND_NORMAL);
>                   temp = expand_widening_mult (mode, op0, op1, target,
>                                                unsignedp, this_optab);
> 
> expand_operands will expand the constant 255 into QImode and return a
> (const_int -1) for op1.  Passing this constant into expand_widening_mult then
> apparently generates a simple negation operation in HImode instead (via
> expand_const_mult) ...
> 
> It seems this code came in here:
> http://gcc.gnu.org/ml/gcc-patches/2010-04/msg01327.html
> Any suggestions how this ought to be handled?

I think we need to pass the narrow mode explicitly.

Richard.

[Bug rtl-optimization/47299] Widening multiply optimization generates bad code

[doko at ubuntu dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

Matthias Klose <doko at ubuntu dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |doko at ubuntu dot com

--- Comment #2 from Matthias Klose <doko at ubuntu dot com> 2011-01-16 03:51:05 UTC ---
4.5/4.6 regression

[Bug rtl-optimization/47299] [4.6 Regression] Widening multiply optimization generates bad code

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011.01.16 15:53:38
            Version|4.5.0                       |4.6.0
   Target Milestone|---                         |4.6.0
            Summary|Widening multiply           |[4.6 Regression] Widening
                   |optimization generates bad  |multiply optimization
                   |code                        |generates bad code
     Ever Confirmed|0                           |1

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> 2011-01-16 15:53:38 UTC ---
It is a 4.6 regression. 4.5 branch is OK.

[Bug rtl-optimization/47299] [4.6 Regression] Widening multiply optimization generates bad code

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1
                 CC|                            |jakub at gcc dot gnu.org

[Bug rtl-optimization/47299] [4.6 Regression] Widening multiply optimization generates bad code

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
         AssignedTo|unassigned at gcc dot       |jakub at gcc dot gnu.org
                   |gnu.org                     |

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-01-17 11:25:00 UTC ---
Created attachment 22994
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22994
gcc46-pr47299.patch

Untested fix.

[Bug rtl-optimization/47299] [4.6 Regression] Widening multiply optimization generates bad code

[doko at ubuntu dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

--- Comment #5 from Matthias Klose <doko at ubuntu dot com> 2011-01-17 23:05:35 UTC ---
a biarch build for yesterday's trunk on i686 with the proposed patch applied
succeeds and doesn't show any regressions.

[Bug rtl-optimization/47299] [4.6 Regression] Widening multiply optimization generates bad code

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-01-18 07:45:17 UTC ---
Author: jakub
Date: Tue Jan 18 07:45:12 2011
New Revision: 168944

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=168944
Log:
    PR rtl-optimization/47299
    * expr.c (expand_expr_real_2) <case WIDEN_MULT_EXPR>: Don't use
    subtarget.  Use normal multiplication if both operands are
    constants.
    * expmed.c (expand_widening_mult): Don't try to optimize constant
    multiplication if op0 has VOIDmode.  Convert op1 constant to mode
    before using it.

    * gcc.c-torture/execute/pr47299.c: New test.

Added:
    trunk/gcc/testsuite/gcc.c-torture/execute/pr47299.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/expmed.c
    trunk/gcc/expr.c
    trunk/gcc/testsuite/ChangeLog

[Bug rtl-optimization/47299] [4.6 Regression] Widening multiply optimization generates bad code

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47299

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

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

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-01-18 08:15:02 UTC ---
Fixed.