This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Patch] Detect true-dependency for strict_low_part in scheduling


>Hi,

>On Mon, 24 Feb 2003, Hartmut Penner wrote:

> > > The function is already in df.c, I think it should simply get
exported
> > > from there (with this clause fixed as well)
> >
> > Exported and changed function 'read_modify_subreg_p' in df.c.
> > Michael had done that last change in that function,


> The version in my copy, which I'll commit once I find time to bootstrap
> the whole change (the last time I tried HEAD wasn't bootstrapping for
> unrelated reasons) looks like so:

Could you also do the changes for df.h (exporting function),
for sched_analyze_1 been able to call it.

> static inline bool
> read_modify_subreg_p (x)
>     rtx x;
> {
>  unsigned int isize, osize;
>  if (GET_CODE (x) != SUBREG)
>    return false;
>  isize = GET_MODE_SIZE (GET_MODE (SUBREG_REG (x)));
>  osize = GET_MODE_SIZE (GET_MODE (x));
>  /* Paradoxical subreg writes don't leave a trace of the old content.  */
>  return (isize > osize && isize > UNITS_PER_WORD);
>}
regards,
      Hartmut


Ciao,
Michael.



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]