Re: [Fwd: Re: Cleanup REG_VTABLE_REF, some old warn_notreached tricks]

[Steven Bosscher <s dot bosscher at student dot tudelft dot nl>]

> Doh, it's not mentioned in this patch, but on cvs you removed
> 
> 
> --- gcc/gcc/final.c     2004/07/15 01:07:49     1.321
> +++ gcc/gcc/final.c     2004/07/17 00:31:06     1.322
> @@ -1999,7 +1999,6 @@
>         rtx body = PATTERN (insn);
>         int insn_code_number;
>         const char *template;
> -       rtx note;
> 
> Which breaks bootstrap on darwin here.
> 
> Accidental ??

Yes.  I broke all targets that use define_peephole and all cc0
targets.  Shame on me! ;)

I'll test the attached patch on powerpc, and with a cross build
for a cc0 target.  You should only need the second hunk for ppc.

Gr.
Steven


Index: final.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/final.c,v
retrieving revision 1.322
diff -u -3 -p -r1.322 final.c
--- final.c     17 Jul 2004 00:31:06 -0000      1.322
+++ final.c     17 Jul 2004 09:08:42 -0000
@@ -2008,17 +2008,19 @@ final_scan_insn (rtx insn, FILE *file, i
          break;

 #ifdef HAVE_cc0
-       /* If there is a REG_CC_SETTER note on this insn, it means that
-          the setting of the condition code was done in the delay slot
-          of the insn that branched here.  So recover the cc status
-          from the insn that set it.  */
+       {
+         /* If there is a REG_CC_SETTER note on this insn, it means that
+            the setting of the condition code was done in the delay slot
+            of the insn that branched here.  So recover the cc status
+            from the insn that set it.  */

-       note = find_reg_note (insn, REG_CC_SETTER, NULL_RTX);
-       if (note)
-         {
-           NOTICE_UPDATE_CC (PATTERN (XEXP (note, 0)), XEXP (note, 0));
-           cc_prev_status = cc_status;
-         }
+         rtx note = find_reg_note (insn, REG_CC_SETTER, NULL_RTX);
+         if (note)
+           {
+             NOTICE_UPDATE_CC (PATTERN (XEXP (note, 0)), XEXP (note, 0));
+             cc_prev_status = cc_status;
+           }
+       }
 #endif

        /* Detect insns that are really jump-tables
@@ -2401,7 +2403,7 @@ final_scan_insn (rtx insn, FILE *file, i
               emit them before the peephole.  */
            if (next != 0 && next != NEXT_INSN (insn))
              {
-               rtx prev = PREV_INSN (insn);
+               rtx note, prev = PREV_INSN (insn);

                for (note = NEXT_INSN (insn); note != next;
                     note = NEXT_INSN (note))