[PATCH] tree-optimization/101173 - fix interchange dependence checking

[Richard Biener]

This adjusts the loop interchange dependence checking to disallow
an outer loop dependence distance of zero.

Bootstrapped and tested on x86_64-unknown-linux-gnu, OK?

Thanks,
Richard.

2021-06-23  Richard Biener  <rguent...@suse.de>

        PR tree-optimization/101173
        * gimple-loop-interchange.cc
        (tree_loop_interchange::valid_data_dependences): Disallow outer
        loop dependence distance of zero.

        * gcc.dg/torture/pr101173.c: New testcase.
---
 gcc/gimple-loop-interchange.cc          |  4 ++--
 gcc/testsuite/gcc.dg/torture/pr101173.c | 18 ++++++++++++++++++
 2 files changed, 20 insertions(+), 2 deletions(-)
 create mode 100644 gcc/testsuite/gcc.dg/torture/pr101173.c

diff --git a/gcc/gimple-loop-interchange.cc b/gcc/gimple-loop-interchange.cc
index 80f749b6071..43045c5455e 100644
--- a/gcc/gimple-loop-interchange.cc
+++ b/gcc/gimple-loop-interchange.cc
@@ -1043,8 +1043,8 @@ tree_loop_interchange::valid_data_dependences (unsigned 
i_idx, unsigned o_idx,
            continue;
 
          /* Be conservative, skip case if either direction at i_idx/o_idx
-            levels is not '=' or '<'.  */
-         if (dist_vect[i_idx] < 0 || dist_vect[o_idx] < 0)
+            levels is not '=' (for the inner loop) or '<'.  */
+         if (dist_vect[i_idx] < 0 || dist_vect[o_idx] <= 0)
            return false;
        }
     }
diff --git a/gcc/testsuite/gcc.dg/torture/pr101173.c 
b/gcc/testsuite/gcc.dg/torture/pr101173.c
new file mode 100644
index 00000000000..0c9090d6686
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/pr101173.c
@@ -0,0 +1,18 @@
+/* { dg-do run } */
+/* { dg-additional-options "-floop-interchange" } */
+
+int a[6][9];
+int main()
+{
+  a[1][3] = 8;
+  for (int b = 1; b <= 5; b++)
+    for (int d = 0; d <= 5; d++)
+#pragma GCC unroll 0
+      for (int c = 0; c <= 5; c++)
+        a[b][c] = a[b][c + 2] & 216;
+  for (int e = 0; e < 6; e++)
+    for (int f = 0; f < 9; f++)
+      if (a[e][f] != 0)
+        __builtin_abort ();
+  return 0;
+}
-- 
2.26.2