vectorizer: Remove dead scalar .COND_* calls from vectorized loops [PR99767]

The following testcase ICEs because disabling of DCE means there are dead
stmts in the loop (though, in theory they could become dead only shortly
before if-conv through some optimization), ifcvt which goes through all
stmts in the loop if-converts them into .COND_DIV etc. internal fn calls
in the copy of the loop meant for vectorization only, the loop is
successfully vectorized but the particular .COND_* call is not because
it isn't a live statement and the scalar .COND_* remains in the IL until
expansion where it ICEs because these ifns only support vectors and not
scalars.

These ifns are similar to .MASK_{LOAD,STORE} in this behavior.

One possible fix could be to expand scalar versions of them during
expansion, basically undoing what if-conv did to create them, i.e.
expand them as the lhs = else; if (mask) { lhs = statement; } or so.

For .MASK_LOAD we have code to replace them in vect_transform_loop already
though (not needed for .MASK_STORE, as stores should be always live
and thus always vectorized), so this patch instead replaces .COND_*
similarly to .MASK_LOAD in that loop, with the small difference
that lhs = .MASK_LOAD (...); is replaced by lhs = 0; while
lhs = .COND_* (..., else_arg); is replaced by lhs = else_arg.
The statement must be dead, otherwise it would be vectorized, so I think
it is not a big deal we don't turn it back into multiple basic blocks etc.
(and it might be not possible to do that at that point).

2021-04-16  Jakub Jelinek  <jakub@redhat.com>

	PR target/99767
	* tree-vect-loop.c (vect_transform_loop): Don't remove just
	dead scalar .MASK_LOAD calls, but also dead .COND_* calls - replace
	them by their last argument.

	* gcc.target/aarch64/pr99767.c: New test.

diff --git a/gcc/testsuite/gcc.target/aarch64/pr99767.c b/gcc/testsuite/gcc.target/aarch64/pr99767.c
new file mode 100644 (file)
index 0000000..5bbfcbd
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/pr99767.c
@@ -0,0 +1,16 @@
+/* PR target/99767 */
+/* { dg-do compile } */
+/* { dg-options " -O1 -fopenmp-simd -fno-tree-dce -march=armv8-a+sve" } */
+
+int a[1024], b[1024];
+
+void
+foo (void)
+{
+  #pragma omp simd
+  for (int i = 0; i < 1024; i++)
+    if (b[i] > 23) {
+      a[i] = b[i] + 1;
+      int v = 1 / 0;   /* { dg-warning "division by zero" } */
+    }
+}

diff --git a/gcc/tree-vect-loop.c b/gcc/tree-vect-loop.c
index 93fa2928e001c154bd4a9a73ac1dbbbf73c456df..2aba503fef7b195c7dd18c3ff6977add1ab73d3c 100644 (file)
--- a/gcc/tree-vect-loop.c
+++ b/gcc/tree-vect-loop.c
@@ -9676,7 +9676,10 @@ vect_transform_loop (loop_vec_info loop_vinfo, gimple *loop_vectorized_call)
           !gsi_end_p (gsi); gsi_next (&gsi))
        {
          gcall *call = dyn_cast <gcall *> (gsi_stmt (gsi));
-         if (call && gimple_call_internal_p (call, IFN_MASK_LOAD))
+         if (!call || !gimple_call_internal_p (call))
+           continue;
+         internal_fn ifn = gimple_call_internal_fn (call);
+         if (ifn == IFN_MASK_LOAD)
            {
              tree lhs = gimple_get_lhs (call);
              if (!VECTOR_TYPE_P (TREE_TYPE (lhs)))
@@ -9686,6 +9689,17 @@ vect_transform_loop (loop_vec_info loop_vinfo, gimple *loop_vectorized_call)
                  gsi_replace (&gsi, new_stmt, true);
                }
            }
+         else if (conditional_internal_fn_code (ifn) != ERROR_MARK)
+           {
+             tree lhs = gimple_get_lhs (call);
+             if (!VECTOR_TYPE_P (TREE_TYPE (lhs)))
+               {
+                 tree else_arg
+                   = gimple_call_arg (call, gimple_call_num_args (call) - 1);
+                 gimple *new_stmt = gimple_build_assign (lhs, else_arg);
+                 gsi_replace (&gsi, new_stmt, true);
+               }
+           }
        }
     }                          /* BBs in loop */