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: Get rid of NOTE_LINE_NUMBER infrastructure


> Sorry if this has been reported before, but...
> 
> Jan Hubicka <jh@suse.cz> writes:
> > *************** print_rtl (FILE *outf, rtx rtx_first)
> > *** 744,751 ****
> >         case CODE_LABEL:
> >         case BARRIER:
> >   	for (tmp_rtx = rtx_first; tmp_rtx != 0; tmp_rtx = NEXT_INSN (tmp_rtx))
> > ! 	  if (! flag_dump_unnumbered
> > ! 	      || !NOTE_P (tmp_rtx) || NOTE_LINE_NUMBER (tmp_rtx) < 0)
> >   	    {
> >   	      fputs (print_rtx_head, outfile);
> >   	      print_rtx (tmp_rtx);
> > --- 725,731 ----
> >         case CODE_LABEL:
> >         case BARRIER:
> >   	for (tmp_rtx = rtx_first; tmp_rtx != 0; tmp_rtx = NEXT_INSN (tmp_rtx))
> > ! 	  if (! flag_dump_unnumbered)
> >   	    {
> >   	      fputs (print_rtx_head, outfile);
> >   	      print_rtx (tmp_rtx);
> 
> ...it looks like you checked in something else:
> 
> [...]
> @@ -743,14 +722,6 @@ print_rtl (FILE *outf, rtx rtx_first)
>        case NOTE:
>        case CODE_LABEL:
>        case BARRIER:
> -	for (tmp_rtx = rtx_first; tmp_rtx != 0; tmp_rtx = NEXT_INSN (tmp_rtx))
> -	  if (! flag_dump_unnumbered
> -	      || !NOTE_P (tmp_rtx) || NOTE_LINE_NUMBER (tmp_rtx) < 0)
> -	    {
> -	      fputs (print_rtx_head, outfile);
> -	      print_rtx (tmp_rtx);
> -	      fprintf (outfile, "\n");
> -	    }
>  	break;
>  
>        default:
> [...]
> 
> This makes print_rtl a no-op in most cases.

OOps, sorry.  There was last minute change in the patch regarding
print_rtl and it seems I've comitted wrong one.  I will commit fix as
obvoius after retesting.

Honza
> 
> Richard


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