diff mbox

Fix PR21273

Message ID 566B2408.8000304@t-online.de
State New
Headers show

Commit Message

Bernd Schmidt Dec. 11, 2015, 7:29 p.m. UTC 
Maybe not the most important PR in the database, but we might as well 
fix and close it. Count the number of alternatives in a MATCH_SCRATCH 
against the max.

Bootstrapped and tested on x86_64-linux (one testcase timed out, almost 
certainly because another test went heavily into swap at one point). Ok?


Bernd

Comments

Jeff Law Dec. 11, 2015, 7:47 p.m. UTC | #1 
On 12/11/2015 12:29 PM, Bernd Schmidt wrote:
> Maybe not the most important PR in the database, but we might as well
> fix and close it. Count the number of alternatives in a MATCH_SCRATCH
> against the max.
>
> Bootstrapped and tested on x86_64-linux (one testcase timed out, almost
> certainly because another test went heavily into swap at one point). Ok?
>
>
> Bernd
>
>
> supscratch.diff
>
>
> 	PR middle-end/21273
> 	* gensupport.c (collect_insn_data): Look for number of alternatives
> 	in MATCH_SCRATCH.
OK.  I haven't been watching closely, but you might be in the running 
for oldest bug fixed in the gcc-6 release :-)

Thanks,
Jeff
diff mbox

Patch

	PR middle-end/21273
	* gensupport.c (collect_insn_data): Look for number of alternatives
	in MATCH_SCRATCH.

Index: gcc/gensupport.c
===================================================================
--- gcc/gensupport.c	(revision 231532)
+++ gcc/gensupport.c	(working copy)
@@ -1068,12 +1068,12 @@  collect_insn_data (rtx pattern, int *pal
   switch (code)
     {
     case MATCH_OPERAND:
-      i = n_alternatives (XSTR (pattern, 2));
+    case MATCH_SCRATCH:
+      i = n_alternatives (XSTR (pattern, code == MATCH_SCRATCH ? 1 : 2));
       *palt = (i > *palt ? i : *palt);
       /* Fall through.  */
 
     case MATCH_OPERATOR:
-    case MATCH_SCRATCH:
     case MATCH_PARALLEL:
       i = XINT (pattern, 0);
       if (i > *pmax)