public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r13-2772] tree-optimization/106922 - missed FRE/PRE
@ 2022-09-22 11:10 Richard Biener
  0 siblings, 0 replies; only message in thread
From: Richard Biener @ 2022-09-22 11:10 UTC (permalink / raw)
  To: gcc-cvs

https://gcc.gnu.org/g:9baee6181b4e427e0b5ba417e51424c15858dce7

commit r13-2772-g9baee6181b4e427e0b5ba417e51424c15858dce7
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Sep 21 13:52:56 2022 +0200

    tree-optimization/106922 - missed FRE/PRE
    
    The following enhances the store-with-same-value trick in
    vn_reference_lookup_3 by not only looking for
    
      a = val;
      *ptr = val;
      .. = a;
    
    but also
    
      *ptr = val;
      other = x;
      .. = a;
    
    where the earlier store is more than one hop away.  It does this
    by queueing the actual value to compare until after the walk but
    as disadvantage only allows a single such skipped store from a
    constant value.
    
    Unfortunately we cannot handle defs from non-constants this way
    since we're prone to pick up values from the past loop iteration
    this way and we have no good way to identify values that are
    invariant in the currently iterated cycle.  That's why we keep
    the single-hop lookup for those cases.  gcc.dg/tree-ssa/pr87126.c
    would be a testcase that's un-XFAILed when we'd handle those
    as well.
    
            PR tree-optimization/106922
            * tree-ssa-sccvn.cc (vn_walk_cb_data::same_val): New member.
            (vn_walk_cb_data::finish): Perform delayed verification of
            a skipped may-alias.
            (vn_reference_lookup_pieces): Likewise.
            (vn_reference_lookup): Likewise.
            (vn_reference_lookup_3): When skipping stores of the same
            value also handle constant stores that are more than a
            single VDEF away by delaying the verification.
    
            * gcc.dg/tree-ssa/ssa-fre-100.c: New testcase.
            * g++.dg/tree-ssa/pr106922.C: Adjust.

Diff:
---
 gcc/testsuite/g++.dg/tree-ssa/pr106922.C    |  3 +-
 gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-100.c | 25 ++++++++
 gcc/tree-ssa-sccvn.cc                       | 97 ++++++++++++++++++++---------
 3 files changed, 93 insertions(+), 32 deletions(-)

diff --git a/gcc/testsuite/g++.dg/tree-ssa/pr106922.C b/gcc/testsuite/g++.dg/tree-ssa/pr106922.C
index faf379b0361..14fa061de20 100644
--- a/gcc/testsuite/g++.dg/tree-ssa/pr106922.C
+++ b/gcc/testsuite/g++.dg/tree-ssa/pr106922.C
@@ -87,5 +87,4 @@ void testfunctionfoo() {
   }
 }
 
-// { dg-final { scan-tree-dump-times "Found fully redundant value" 4 "pre" { xfail { ! lp64 } } } }
-// { dg-final { scan-tree-dump-not "m_initialized" "cddce3" { xfail { ! lp64 } } } }
+// { dg-final { scan-tree-dump-not "m_initialized" "dce3" } }
diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-100.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-100.c
new file mode 100644
index 00000000000..ead76548f3d
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-100.c
@@ -0,0 +1,25 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -fdump-tree-fre1" } */
+
+float bar, baz;
+void foo (int *p, int n)
+{
+  *p = 0;
+  do
+    {
+      bar = 1.;
+      /* When iterating we should have optimistically value-numbered
+	 *p to zero, on the second iteration we have to prove the
+	 store below does not affect the value of this load though.
+	 We can compare the stored value against the value from the
+	 previous iteration instead relying on a non-walking lookup.  */
+      if (*p)
+        {
+          baz = 2.;
+          *p = 0;
+        }
+    }
+  while (--n);
+}
+
+/* { dg-final { scan-tree-dump-not "baz" "fre1" } } */
diff --git a/gcc/tree-ssa-sccvn.cc b/gcc/tree-ssa-sccvn.cc
index 85a7698f694..9c12a8e4f03 100644
--- a/gcc/tree-ssa-sccvn.cc
+++ b/gcc/tree-ssa-sccvn.cc
@@ -1803,7 +1803,8 @@ struct vn_walk_cb_data
 		   vn_lookup_kind vn_walk_kind_, bool tbaa_p_, tree mask_,
 		   bool redundant_store_removal_p_)
     : vr (vr_), last_vuse_ptr (last_vuse_ptr_), last_vuse (NULL_TREE),
-      mask (mask_), masked_result (NULL_TREE), vn_walk_kind (vn_walk_kind_),
+      mask (mask_), masked_result (NULL_TREE), same_val (NULL_TREE),
+      vn_walk_kind (vn_walk_kind_),
       tbaa_p (tbaa_p_), redundant_store_removal_p (redundant_store_removal_p_),
       saved_operands (vNULL), first_set (-2), first_base_set (-2),
       known_ranges (NULL)
@@ -1864,6 +1865,7 @@ struct vn_walk_cb_data
   tree last_vuse;
   tree mask;
   tree masked_result;
+  tree same_val;
   vn_lookup_kind vn_walk_kind;
   bool tbaa_p;
   bool redundant_store_removal_p;
@@ -1902,6 +1904,8 @@ vn_walk_cb_data::finish (alias_set_type set, alias_set_type base_set, tree val)
       masked_result = val;
       return (void *) -1;
     }
+  if (same_val && !operand_equal_p (val, same_val))
+    return (void *) -1;
   vec<vn_reference_op_s> &operands
     = saved_operands.exists () ? saved_operands : vr->operands;
   return vn_reference_lookup_or_insert_for_pieces (last_vuse, set, base_set,
@@ -2675,36 +2679,57 @@ vn_reference_lookup_3 (ao_ref *ref, tree vuse, void *data_,
 	 and return the found value.  */
       if (is_gimple_reg_type (TREE_TYPE (lhs))
 	  && types_compatible_p (TREE_TYPE (lhs), vr->type)
-	  && (ref->ref || data->orig_ref.ref))
-	{
-	  tree *saved_last_vuse_ptr = data->last_vuse_ptr;
-	  /* Do not update last_vuse_ptr in vn_reference_lookup_2.  */
-	  data->last_vuse_ptr = NULL;
-	  tree saved_vuse = vr->vuse;
-	  hashval_t saved_hashcode = vr->hashcode;
-	  void *res = vn_reference_lookup_2 (ref, gimple_vuse (def_stmt), data);
-	  /* Need to restore vr->vuse and vr->hashcode.  */
-	  vr->vuse = saved_vuse;
-	  vr->hashcode = saved_hashcode;
-	  data->last_vuse_ptr = saved_last_vuse_ptr;
-	  if (res && res != (void *)-1)
+	  && (ref->ref || data->orig_ref.ref)
+	  && !data->same_val
+	  && !data->mask
+	  && data->partial_defs.is_empty ()
+	  && multiple_p (get_object_alignment
+			   (ref->ref ? ref->ref : data->orig_ref.ref),
+			   ref->size)
+	  && multiple_p (get_object_alignment (lhs), ref->size))
+	{
+	  tree rhs = gimple_assign_rhs1 (def_stmt);
+	  /* ???  We may not compare to ahead values which might be from
+	     a different loop iteration but only to loop invariants.  Use
+	     CONSTANT_CLASS_P (unvalueized!) as conservative approximation.
+	     The one-hop lookup below doesn't have this issue since there's
+	     a virtual PHI before we ever reach a backedge to cross.  */
+	  if (CONSTANT_CLASS_P (rhs))
 	    {
-	      vn_reference_t vnresult = (vn_reference_t) res;
-	      tree rhs = gimple_assign_rhs1 (def_stmt);
-	      if (TREE_CODE (rhs) == SSA_NAME)
-		rhs = SSA_VAL (rhs);
-	      if (vnresult->result
-		  && operand_equal_p (vnresult->result, rhs, 0)
-		  /* We have to honor our promise about union type punning
-		     and also support arbitrary overlaps with
-		     -fno-strict-aliasing.  So simply resort to alignment to
-		     rule out overlaps.  Do this check last because it is
-		     quite expensive compared to the hash-lookup above.  */
-		  && multiple_p (get_object_alignment
-				   (ref->ref ? ref->ref : data->orig_ref.ref),
-				 ref->size)
-		  && multiple_p (get_object_alignment (lhs), ref->size))
-		return res;
+	      if (dump_file && (dump_flags & TDF_DETAILS))
+		{
+		  fprintf (dump_file,
+			   "Skipping possible redundant definition ");
+		  print_gimple_stmt (dump_file, def_stmt, 0);
+		}
+	      /* Delay the actual compare of the values to the end of the walk
+		 but do not update last_vuse from here.  */
+	      data->last_vuse_ptr = NULL;
+	      data->same_val = rhs;
+	      return NULL;
+	    }
+	  else
+	    {
+	      tree *saved_last_vuse_ptr = data->last_vuse_ptr;
+	      /* Do not update last_vuse_ptr in vn_reference_lookup_2.  */
+	      data->last_vuse_ptr = NULL;
+	      tree saved_vuse = vr->vuse;
+	      hashval_t saved_hashcode = vr->hashcode;
+	      void *res = vn_reference_lookup_2 (ref, gimple_vuse (def_stmt),
+						 data);
+	      /* Need to restore vr->vuse and vr->hashcode.  */
+	      vr->vuse = saved_vuse;
+	      vr->hashcode = saved_hashcode;
+	      data->last_vuse_ptr = saved_last_vuse_ptr;
+	      if (res && res != (void *)-1)
+		{
+		  vn_reference_t vnresult = (vn_reference_t) res;
+		  if (TREE_CODE (rhs) == SSA_NAME)
+		    rhs = SSA_VAL (rhs);
+		  if (vnresult->result
+		      && operand_equal_p (vnresult->result, rhs, 0))
+		    return res;
+		}
 	    }
 	}
     }
@@ -3798,6 +3823,14 @@ vn_reference_lookup_pieces (tree vuse, alias_set_type set,
       if (ops_for_ref != shared_lookup_references)
 	ops_for_ref.release ();
       gcc_checking_assert (vr1.operands == shared_lookup_references);
+      if (*vnresult
+	  && data.same_val
+	  && (!(*vnresult)->result
+	      || !operand_equal_p ((*vnresult)->result, data.same_val)))
+	{
+	  *vnresult = NULL;
+	  return NULL_TREE;
+	}
     }
 
   if (*vnresult)
@@ -3913,6 +3946,10 @@ vn_reference_lookup (tree op, tree vuse, vn_lookup_kind kind,
       if (wvnresult)
 	{
 	  gcc_assert (mask == NULL_TREE);
+	  if (data.same_val
+	      && (!wvnresult->result
+		  || !operand_equal_p (wvnresult->result, data.same_val)))
+	    return NULL_TREE;
 	  if (vnresult)
 	    *vnresult = wvnresult;
 	  return wvnresult->result;

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2022-09-22 11:10 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-09-22 11:10 [gcc r13-2772] tree-optimization/106922 - missed FRE/PRE Richard Biener

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).