Patchwork Don't bypass blocks with multiple latch edges (PR middle-end/54838)

login
register
mail settings
Submitter Marek Polacek
Date Nov. 30, 2012, 4:16 p.m.
Message ID <20121130161654.GG10621@redhat.com>
Download mbox | patch
Permalink /patch/202983/
State New
Headers show

Comments

Marek Polacek - Nov. 30, 2012, 4:16 p.m.
On Fri, Nov 30, 2012 at 10:01:37AM +0100, Richard Biener wrote:
> RTL cprop seems to run both before and after RTL loop optimizers (currently
> after RTL loop optimizers we throw away loops - an arbitrary chosen point
> before IRA across which I could not get things to work).  Thus you could do
> 
>   if (current_loops)
>     is_loop_header = bb == bb->loop_father->header;
>   else
>     {
>   may_be_loop_header = false;
>   FOR_EACH_EDGE (e, ei, bb->preds)
>     if (e->flags & EDGE_DFS_BACK)
>       {
>         may_be_loop_header = true;
>         break;
>       }
>     }

I can do this as a followup.
 
> I don't understand
> 
>       /* The irreducible loops created by redirecting of edges entering the
>          loop from outside would decrease effectiveness of some of the
>          following optimizations, so prevent this.  */
>       if (may_be_loop_header
>           && !(e->flags & EDGE_DFS_BACK))
>         {
>           ei_next (&ei);
>           continue;
>         }
> 
> why isn't this simply
> 
>       if (may_be_loop_header)
>         {
>           ei_next (&ei);
>           continue;
>         }
> 
> ?  It looks like the code tries to allow "rotating" a loop - but that's only
> good if bb has exactly two predecessors (one entry and one latch edge).
> And even then it requires to manually update the loop structures (update
> what the new header and latch blocks are).
> 
> That said, removing the !(e->flags & EDGE_DFS_BACK) condition seems
> to fix the ICE.  Threading across a loop header is in fact complicated
> (see the special routine tree-ssa-threadupdate.c:thread_through_loop_header
> necessary for that).  Let's declare the GIMPLE level did all interesting
> threadings through headers.

Agreed.  This is the fix I had some time ago, but at that time it
didn't seem like such a great idea.  Done this time around.
Regtested/bootstrapped on x86_64-linux, ok for trunk?

2012-11-30  Marek Polacek  <polacek@redhat.com>

	PR middle-end/54838
	* cprop.c (bypass_block): Skip header edges.

	* gcc.dg/pr54838.c: New test.


	Marek

Patch

--- gcc/cprop.c.mp	2012-11-29 15:49:53.120524295 +0100
+++ gcc/cprop.c	2012-11-30 10:30:23.509501957 +0100
@@ -1539,8 +1539,7 @@  bypass_block (basic_block bb, rtx setcc,
       /* The irreducible loops created by redirecting of edges entering the
 	 loop from outside would decrease effectiveness of some of the
 	 following optimizations, so prevent this.  */
-      if (may_be_loop_header
-	  && !(e->flags & EDGE_DFS_BACK))
+      if (may_be_loop_header)
 	{
 	  ei_next (&ei);
 	  continue;
--- gcc/testsuite/gcc.dg/pr54838.c.mp	2012-11-26 14:48:43.783980854 +0100
+++ gcc/testsuite/gcc.dg/pr54838.c	2012-11-29 17:43:19.397737779 +0100
@@ -0,0 +1,24 @@ 
+/* PR middle-end/54838 */
+/* { dg-do compile } */
+/* { dg-options "-O2 -fno-forward-propagate -ftracer" } */
+
+void bar (void);
+
+void
+foo (void *b, int *c)
+{
+again:
+  switch (*c)
+    {
+    case 1:
+      if (!b)
+	{
+	  bar ();
+	  return;
+	}
+      goto again;
+    case 3:
+      if (!b)
+	goto again;
+    }
+}