[PATCH] Refactor is_non_loop_exit_postdominating

[PATCH] Refactor is_non_loop_exit_postdominating

[Richard Biener]

That's a weird function in predicate analysis that currently looks like

/* Return true if BB1 is postdominating BB2 and BB1 is not a loop exit
   bb.  The loop exit bb check is simple and does not cover all cases.  */
static bool
is_non_loop_exit_postdominating (basic_block bb1, basic_block bb2)
{
  if (!dominated_by_p (CDI_POST_DOMINATORS, bb2, bb1))
    return false;
  if (single_pred_p (bb1) && !single_succ_p (bb2))
    return false;
  return true;
}

One can refactor this to

  return (dominated_by_p (CDI_POST_DOMINATORS, bb2, bb1)
          && !(single_pred_p (bb1) && !single_succ_p (bb2)));

Notable is that the comment refers to BB1 with respect to a loop
exit but the test seems to be written with an exit edge bb1 -> bb2
in mind.  From the three callers only a single one is guaranteed
to have bb1 and bb2 connected directly with an edge.

The patch now introduces a is_loop_exit function and inlines
the post-dominance check which makes the find_control_equiv_block
case simpler because the post-dominance check can be elided.
It also avoids the double negation in compute_control_dep_chain
and makes it obvious this is the case where we do look at an edge.
For the main is_use_guarded API I chose to elide the loop exit
test, if the use block post-dominates the definition block of the
PHI node the use is always unconditional.  I don't quite understand
the loop exit special-casing of the remaining two uses though.

Bootstrap and regtest running on x86_64-unknown-linux-gnu.

The code is this way since the original r0-99656-g34f97b9407c6a3
so I'm not hoping for any insights ... :/

	* gimple-predicate-analysis.cc (is_loop_exit): Split out
	from ...
	(is_non_loop_exit_postdominating): ... here.  Remove after
	inlining ...
	(find_control_equiv_block): ... here.
	(compute_control_dep_chain): ... and here.
	(predicate::is_use_guarded): Do not excempt loop exits
	from short-cutting the case of the use post-dominating the
	PHI definition.
---
 gcc/gimple-predicate-analysis.cc | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/gcc/gimple-predicate-analysis.cc b/gcc/gimple-predicate-analysis.cc
index d1b0d1283dc..4cf05951ec7 100644
--- a/gcc/gimple-predicate-analysis.cc
+++ b/gcc/gimple-predicate-analysis.cc
@@ -46,19 +46,12 @@
 
 #define DEBUG_PREDICATE_ANALYZER 1
 
-/* Return true if BB1 is postdominating BB2 and BB1 is not a loop exit
-   bb.  The loop exit bb check is simple and does not cover all cases.  */
+/* Return true if, when BB1 is postdominating BB2, BB1 is a loop exit.  */
 
 static bool
-is_non_loop_exit_postdominating (basic_block bb1, basic_block bb2)
+is_loop_exit (basic_block bb2, basic_block bb1)
 {
-  if (!dominated_by_p (CDI_POST_DOMINATORS, bb2, bb1))
-    return false;
-
-  if (single_pred_p (bb1) && !single_succ_p (bb2))
-    return false;
-
-  return true;
+  return single_pred_p (bb1) && !single_succ_p (bb2);
 }
 
 /* Find BB's closest postdominator that is its control equivalent (i.e.,
@@ -70,7 +63,7 @@ find_control_equiv_block (basic_block bb)
   basic_block pdom = get_immediate_dominator (CDI_POST_DOMINATORS, bb);
 
   /* Skip the postdominating bb that is also a loop exit.  */
-  if (!is_non_loop_exit_postdominating (pdom, bb))
+  if (is_loop_exit (bb, pdom))
     return NULL;
 
   /* If the postdominator is dominated by BB, return it.  */
@@ -1112,10 +1105,11 @@ compute_control_dep_chain (basic_block dom_bb, const_basic_block dep_bb,
       if (e->flags & (EDGE_FAKE | EDGE_ABNORMAL))
 	continue;
 
-      basic_block cd_bb = e->dest;
       cur_cd_chain.safe_push (e);
-      while (!is_non_loop_exit_postdominating (cd_bb, dom_bb))
+      while (!dominated_by_p (CDI_POST_DOMINATORS, e->src, e->dest)
+	     || is_loop_exit (e->src, e->dest))
 	{
+	  basic_block cd_bb = e->dest;
 	  if (cd_bb == dep_bb)
 	    {
 	      /* Found a direct control dependence.  */
@@ -1885,7 +1879,7 @@ predicate::is_use_guarded (gimple *use_stmt, basic_block use_bb,
      in the same bb.  */
   predicate use_preds (def_bb, use_bb, m_eval);
 
-  if (is_non_loop_exit_postdominating (use_bb, def_bb))
+  if (dominated_by_p (CDI_POST_DOMINATORS, def_bb, use_bb))
     {
       if (is_empty ())
 	{
-- 
2.35.3