[Bug target/48970] New: Inaccurate comments for processor_costs

[Bug target/48970] New: Inaccurate comments for processor_costs

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48970

           Summary: Inaccurate comments for processor_costs
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: hjl.tools@gmail.com
                CC: ubizjak@gmail.com
            Target: x86


i386.h has

struct processor_costs {
...
  const int sse_load[3];        /* cost of loading SSE register
                                   in SImode, DImode and TImode*/
  const int sse_store[3];       /* cost of storing SSE register
                                   in SImode, DImode and TImode*/

But i386.c has
 if (SSE_CLASS_P (regclass))
    {
      int index;
      switch (GET_MODE_SIZE (mode))
        {     
          case 4:
            index = 0;
            break;
          case 8:
            index = 1;
            break;
          case 16:
            index = 2;
            break;
          default:
            return 100;
        }     
      if (in == 2) 
        return MAX (ix86_cost->sse_load [index], ix86_cost->sse_store [index]);
      return in ? ix86_cost->sse_load [index] : ix86_cost->sse_store [index];
    }

SImode, DImode nor TImode aren't used to compute costs.

[Bug target/48970] Inaccurate comments for processor_costs

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48970

--- Comment #1 from H.J. Lu <hjl.tools at gmail dot com> 2011-05-11 21:23:45 UTC ---
Also

 const int mmxsse_to_integer;  /* cost of moving mmxsse register to
                                   integer and vice versa.  */

isn't limited to integers:

 /* Moves between SSE/MMX and integer unit are expensive.  */
  if (MMX_CLASS_P (class1) != MMX_CLASS_P (class2)
      || SSE_CLASS_P (class1) != SSE_CLASS_P (class2))

    /* ??? By keeping returned value relatively high, we limit the number
       of moves between integer and MMX/SSE registers for all targets.
       Additionally, high value prevents problem with x86_modes_tieable_p(),
       where integer modes in MMX/SSE registers are not tieable
       because of missing QImode and HImode moves to, from or between
       MMX/SSE registers.  */
    return MAX (8, ix86_cost->mmxsse_to_integer);

Integer register classes aren't checked at all.

[Bug target/48970] Inaccurate comments for processor_costs

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48970

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
      const int sse_load[5];    /* cost of loading SSE register
                                   in 32bit, 64bit, 128bit, 256bit and 512bit
*/
      const int sse_store[5];   /* cost of storing SSE register
                                   in SImode, DImode and TImode.  */





(In reply to H.J. Lu from comment #1)
> Also
> 
>  const int mmxsse_to_integer;  /* cost of moving mmxsse register to
>                                    integer and vice versa.  */

This changed with PR 90877.