public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [PATCH] tree-optimization/103237 - avoid vectorizing unhandled double reductions
@ 2021-11-15 12:07 Richard Biener
  0 siblings, 0 replies; only message in thread
From: Richard Biener @ 2021-11-15 12:07 UTC (permalink / raw)
  To: gcc-patches

Double reductions which have multiple LC PHIs in the inner loop
are not handled correctly during transformation since those PHIs
are not properly classified as reduction.  The following disables
vectorizing them.

Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.

2021-11-15  Richard Biener  <rguenther@suse.de>

	PR tree-optimization/103237
	* tree-vect-loop.c (vect_is_simple_reduction): Fail for
	double reductions with multiple inner loop LC PHI nodes.

	* gcc.dg/torture/pr103237.c: New testcase.
---
 gcc/testsuite/gcc.dg/torture/pr103237.c | 24 ++++++++++++++++++++++++
 gcc/tree-vect-loop.c                    | 11 +++++++++++
 2 files changed, 35 insertions(+)
 create mode 100644 gcc/testsuite/gcc.dg/torture/pr103237.c

diff --git a/gcc/testsuite/gcc.dg/torture/pr103237.c b/gcc/testsuite/gcc.dg/torture/pr103237.c
new file mode 100644
index 00000000000..f2399f9586e
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/pr103237.c
@@ -0,0 +1,24 @@
+/* { dg-do run } */
+/* { dg-additional-options "-ftree-vectorize" } */
+
+int g1;
+unsigned int g2 = -1U;
+static void __attribute__((noipa))
+func_1()
+{
+  int *l_1 = &g1;
+  for (int g3a = 0; g3a != 4; g3a++)
+    for (int l_2 = 0; l_2 <= 3; l_2++)
+      {
+        unsigned int *l_3 = &g2;
+        *l_1 = *l_3 ^= 1;
+      }
+}
+int
+main()
+{
+  func_1();
+  if (g1 != -1)
+    __builtin_abort ();
+  return 0;
+}
diff --git a/gcc/tree-vect-loop.c b/gcc/tree-vect-loop.c
index 1cd5dbcb6f7..73efdb96bad 100644
--- a/gcc/tree-vect-loop.c
+++ b/gcc/tree-vect-loop.c
@@ -3567,6 +3567,17 @@ vect_is_simple_reduction (loop_vec_info loop_info, stmt_vec_info phi_info,
       return def_stmt_info;
     }
 
+  /* When the inner loop of a double reduction ends up with more than
+     one loop-closed PHI we have failed to classify alternate such
+     PHIs as double reduction, leading to wrong code.  See PR103237.  */
+  if (inner_loop_of_double_reduc && lcphis.length () != 1)
+    {
+      if (dump_enabled_p ())
+	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
+			 "unhandle double reduction\n");
+      return NULL;
+    }
+
   /* If this isn't a nested cycle or if the nested cycle reduction value
      is used ouside of the inner loop we cannot handle uses of the reduction
      value.  */
-- 
2.31.1

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

only message in thread, other threads:[~2021-11-15 12:07 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-11-15 12:07 [PATCH] tree-optimization/103237 - avoid vectorizing unhandled double reductions 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).