public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [PATCH] tree-optimization/107407 - wrong code with DSE
@ 2022-10-28 13:22 Richard Biener
  0 siblings, 0 replies; only message in thread
From: Richard Biener @ 2022-10-28 13:22 UTC (permalink / raw)
  To: gcc-patches

So what happens is that we elide processing of this check with

          /* In addition to kills we can remove defs whose only use
             is another def in defs.  That can only ever be PHIs of which
             we track two for simplicity reasons, the first and last in
             {first,last}_phi_def (we fail for multiple PHIs anyways).
             We can also ignore defs that feed only into
             already visited PHIs.  */
          else if (single_imm_use (vdef, &use_p, &use_stmt)
                   && (use_stmt == first_phi_def
                       || use_stmt == last_phi_def
                       || (gimple_code (use_stmt) == GIMPLE_PHI
                           && bitmap_bit_p (visited,
                                            SSA_NAME_VERSION
                                              (PHI_RESULT (use_stmt))))))

where we have the last PHI being the outer loop virtual PHI and the first
PHI being the loop exit PHI of the outer loop and we've already processed
the single immediate use of the outer loop PHI, the inner loop PHI.  But
we still have to perform the above check!

It's easiest to perform the check when we visit the PHI node instead of
delaying it to the actual processing loop.

Bootstrap & regtest in progress on x86_64-unknown-linux-gnu.

	PR tree-optimization/107407
	* tree-ssa-dse.cc (dse_classify_store): Perform backedge
	varying index check when collecting PHI uses rather than
	after optimizing processing of the candidate defs.

	* gcc.dg/torture/pr107407.c: New testcase.
---
 gcc/testsuite/gcc.dg/torture/pr107407.c | 26 +++++++++++++++++++++++++
 gcc/tree-ssa-dse.cc                     | 17 ++++++++--------
 2 files changed, 35 insertions(+), 8 deletions(-)
 create mode 100644 gcc/testsuite/gcc.dg/torture/pr107407.c

diff --git a/gcc/testsuite/gcc.dg/torture/pr107407.c b/gcc/testsuite/gcc.dg/torture/pr107407.c
new file mode 100644
index 00000000000..228fce1e699
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/pr107407.c
@@ -0,0 +1,26 @@
+/* { dg-do run } */
+
+int *a;
+int c[4];
+int d;
+
+static int
+f(char k, int j)
+{
+  for (; k <= 3; k++)
+    {
+      a = &c[k];
+      for (; d <= 1; d++)
+        *a = 3;
+    }
+  *a = 0;
+}
+
+int main()
+{
+  int i;
+  f(0, 0);
+  if (c[0] != 3)
+    __builtin_abort ();
+  return 0;
+}
diff --git a/gcc/tree-ssa-dse.cc b/gcc/tree-ssa-dse.cc
index c14e5e43eb3..82976bdbeb0 100644
--- a/gcc/tree-ssa-dse.cc
+++ b/gcc/tree-ssa-dse.cc
@@ -978,14 +978,6 @@ dse_classify_store (ao_ref *ref, gimple *stmt,
 
       if (gimple_code (temp) == GIMPLE_PHI)
 	{
-	  /* If we visit this PHI by following a backedge then we have to
-	     make sure ref->ref only refers to SSA names that are invariant
-	     with respect to the loop represented by this PHI node.  */
-	  if (dominated_by_p (CDI_DOMINATORS, gimple_bb (stmt),
-			      gimple_bb (temp))
-	      && !for_each_index (ref->ref ? &ref->ref : &ref->base,
-				  check_name, gimple_bb (temp)))
-	    return DSE_STORE_LIVE;
 	  defvar = PHI_RESULT (temp);
 	  bitmap_set_bit (visited, SSA_NAME_VERSION (defvar));
 	}
@@ -1019,6 +1011,15 @@ dse_classify_store (ao_ref *ref, gimple *stmt,
 	      if (!bitmap_bit_p (visited,
 				 SSA_NAME_VERSION (PHI_RESULT (use_stmt))))
 		{
+		  /* If we visit this PHI by following a backedge then we have
+		     to make sure ref->ref only refers to SSA names that are
+		     invariant with respect to the loop represented by this
+		     PHI node.  */
+		  if (dominated_by_p (CDI_DOMINATORS, gimple_bb (stmt),
+				      gimple_bb (use_stmt))
+		      && !for_each_index (ref->ref ? &ref->ref : &ref->base,
+					  check_name, gimple_bb (use_stmt)))
+		    return DSE_STORE_LIVE;
 		  defs.safe_push (use_stmt);
 		  if (!first_phi_def)
 		    first_phi_def = as_a <gphi *> (use_stmt);
-- 
2.35.3

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

only message in thread, other threads:[~2022-10-28 13:22 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-28 13:22 [PATCH] tree-optimization/107407 - wrong code with DSE 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).